diff --git a/ndarray/src/main/java/org/tensorflow/ndarray/NdArray.java b/ndarray/src/main/java/org/tensorflow/ndarray/NdArray.java
index 5fe51121b13..dfabc0be8c7 100644
--- a/ndarray/src/main/java/org/tensorflow/ndarray/NdArray.java
+++ b/ndarray/src/main/java/org/tensorflow/ndarray/NdArray.java
@@ -55,29 +55,7 @@
  *
  * @param <T> the type of values to be mapped
  */
-public interface NdArray<T> {
-
-  /**
-   * @return the shape of this N-dimensional array
-   */
-  Shape shape();
-
-  /**
-   * @return the rank of this N-dimensional array
-   */
-  default int rank() {
-    return shape().numDimensions();
-  }
-
-  /**
-   * Computes and returns the total size of this N-dimensional array, in number of values.
-   *
-   * <p>For example, given a 3x3x2 matrix, the return value will be 18.
-   * @return total size of this nd array
-   */
-  default long size() {
-    return shape().size();
-  }
+public interface NdArray<T> extends NdArrayBase {
 
   /**
    * Returns a sequence of all elements at a given dimension.
diff --git a/ndarray/src/main/java/org/tensorflow/ndarray/NdArrayBase.java b/ndarray/src/main/java/org/tensorflow/ndarray/NdArrayBase.java
new file mode 100644
index 00000000000..75112ce2470
--- /dev/null
+++ b/ndarray/src/main/java/org/tensorflow/ndarray/NdArrayBase.java
@@ -0,0 +1,47 @@
+/*
+ Copyright 2019 The TensorFlow Authors. All Rights Reserved.
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+     http://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
+ =======================================================================
+ */
+package org.tensorflow.ndarray;
+
+import java.util.function.BiConsumer;
+import java.util.function.Consumer;
+import org.tensorflow.ndarray.buffer.DataBuffer;
+import org.tensorflow.ndarray.index.Index;
+
+public interface NdArrayBase {
+
+  /**
+   * @return the shape of this N-dimensional array
+   */
+  Shape shape();
+
+  /**
+   * @return the rank of this N-dimensional array
+   */
+  default int rank() {
+    return shape().numDimensions();
+  }
+
+  /**
+   * Computes and returns the total size of this N-dimensional array, in number of values.
+   *
+   * <p>For example, given a 3x3x2 matrix, the return value will be 18.
+   * @return total size of this nd array
+   */
+  default long size() {
+    return shape().size();
+  }
+}
diff --git a/ndarray/src/main/java/org/tensorflow/ndarray/NdArrays.java b/ndarray/src/main/java/org/tensorflow/ndarray/NdArrays.java
index 21b33402e98..8ad55cae7ed 100644
--- a/ndarray/src/main/java/org/tensorflow/ndarray/NdArrays.java
+++ b/ndarray/src/main/java/org/tensorflow/ndarray/NdArrays.java
@@ -65,7 +65,7 @@ public static ByteNdArray vectorOf(byte... values) {
     if (values == null) {
       throw new IllegalArgumentException("Values cannot be null");
     }
-    return wrap(DataBuffers.of(values, false, false), Shape.of(values.length));
+    return wrap(Shape.of(values.length), DataBuffers.of(values, false, false));
   }
 
   /**
@@ -81,19 +81,19 @@ public static ByteNdArray ofBytes(Shape shape) {
     if (shape == null) {
       throw new IllegalArgumentException("Shape cannot be null");
     }
-    return wrap(DataBuffers.ofBytes(shape.size()), shape);
+    return wrap(shape, DataBuffers.ofBytes(shape.size()));
   }
 
   /**
    * Wraps a buffer in a byte N-dimensional array of a given shape.
    *
-   * @param buffer buffer to wrap
    * @param shape shape of the array
+   * @param buffer buffer to wrap
    * @return new byte N-dimensional array
    * @throws IllegalArgumentException if shape is null, has unknown dimensions or has size bigger
    *                                  in the buffer size
    */
-  public static ByteNdArray wrap(ByteDataBuffer buffer, Shape shape) {
+  public static ByteNdArray wrap(Shape shape, ByteDataBuffer buffer) {
     return ByteDenseNdArray.create(buffer, shape);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/java_defs.h b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/java_defs.h
index e41dc2dd9df..6028c4ea71d 100644
--- a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/java_defs.h
+++ b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/java_defs.h
@@ -97,9 +97,6 @@ class Type {
   static Type IterableOf(const Type& type) {
     return Interface("Iterable").add_parameter(type);
   }
-  static Type DataTypeOf(const Type& type) {
-    return Class("DataType", "org.tensorflow").add_parameter(type);
-  }
   static Type ForDataType(DataType data_type) {
     switch (data_type) {
       case DataType::DT_BOOL:
diff --git a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_generator.cc b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_generator.cc
index 843f3bdb247..2bec80a8457 100644
--- a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_generator.cc
+++ b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_generator.cc
@@ -177,7 +177,7 @@ void RenderSecondaryFactoryMethod(const OpSpec& op, const Type& op_class,
     if (attr.type().kind() == Type::GENERIC &&
         default_types.find(attr.type().name()) != default_types.end()) {
       factory_statement << default_types.at(attr.type().name()).name()
-                        << ".DTYPE";
+                        << ".class";
     } else {
       AddArgument(attr.var(), attr.description(), &factory, &factory_doc);
       factory_statement << attr.var().name();
diff --git a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_specs.cc b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_specs.cc
index c9e0525edb7..8f676deeead 100644
--- a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_specs.cc
+++ b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/op_specs.cc
@@ -81,13 +81,19 @@ class TypeResolver {
   std::pair<Type, Type> MakeTypePair(const Type& type) {
     return std::make_pair(type, type);
   }
-  Type NextGeneric() {
+  Type NextGeneric(const OpDef_AttrDef& attr_def) {
     char generic_letter = next_generic_letter_++;
     if (next_generic_letter_ > 'Z') {
       next_generic_letter_ = 'A';
     }
-    return Type::Generic(string(1, generic_letter))
-        .add_supertype(Type::Class("TType", "org.tensorflow.types.family"));
+    return Type::Generic(string(1, generic_letter));
+  }
+  Type TypeFamilyOf(const OpDef_AttrDef& attr_def) {
+    // TODO(karllessard) support more type families
+    if (IsRealNumbers(attr_def.allowed_values())) {
+      return Type::Interface("TNumber", "org.tensorflow.types.family");
+    }
+    return Type::Interface("TType", "org.tensorflow.types.family");
   }
 };
 
@@ -152,15 +158,12 @@ std::pair<Type, Type> TypeResolver::TypesOf(const OpDef_AttrDef& attr_def,
     types = MakeTypePair(Type::Class("Shape", "org.tensorflow.ndarray"));
 
   } else if (attr_type == "tensor") {
-    types = MakeTypePair(Type::Class("Tensor", "org.tensorflow")
-                             .add_parameter(Type::Wildcard()));
+    types = MakeTypePair(Type::Class("TType", "org.tensorflow.types.family"));
 
   } else if (attr_type == "type") {
-    Type type = *iterable_out ? Type::Wildcard() : NextGeneric();
-    if (IsRealNumbers(attr_def.allowed_values())) {
-      type.add_supertype(Type::Class("TNumber", "org.tensorflow.types.family"));
-    }
-    types = MakeTypePair(type, Type::Enum("DataType", "org.tensorflow"));
+    Type type = *iterable_out ? Type::Wildcard() : NextGeneric(attr_def);
+    type.add_supertype(TypeFamilyOf(attr_def));
+    types = MakeTypePair(type, Type::Class("Class"));
 
   } else {
     LOG(FATAL) << "Cannot resolve data type for attribute \"" << attr_type
@@ -306,7 +309,7 @@ AttributeSpec CreateAttribute(const OpDef_AttrDef& attr_def,
   bool iterable = false;
   std::pair<Type, Type> types = type_resolver->TypesOf(attr_def, &iterable);
   Type var_type = types.first.kind() == Type::GENERIC
-                      ? Type::DataTypeOf(types.first)
+                      ? Type::ClassOf(types.first)
                       : types.first;
   if (iterable) {
     var_type = Type::ListOf(var_type);
diff --git a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.cc b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.cc
index 8598b1d945d..37315f0dff3 100644
--- a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.cc
+++ b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.cc
@@ -85,6 +85,7 @@ SourceWriter& SourceWriter::Append(const StringPiece& str) {
 SourceWriter& SourceWriter::AppendType(const Type& type) {
   if (type.wildcard()) {
     Append("?");
+    WriteTypeBounds(type.supertypes());
   } else {
     Append(type.name());
     if (!type.parameters().empty()) {
@@ -321,14 +322,27 @@ SourceWriter& SourceWriter::WriteGenerics(
       Append(", ");
     }
     Append(pt->name());
-    if (!pt->supertypes().empty()) {
-      Append(" extends ").AppendType(pt->supertypes().front());
-    }
+    WriteTypeBounds(pt->supertypes());
     first = false;
   }
   return Append(">");
 }
 
+SourceWriter& SourceWriter::WriteTypeBounds(
+    const std::list<Type>& bounds) {
+  bool first = true;
+  for (const Type& bound : bounds) {
+    if (first) {
+      Append(" extends ");
+      first = false;
+    } else {
+      Append(" & ");
+    }
+    AppendType(bound);
+  }
+  return *this;
+}
+
 SourceWriter::GenericNamespace* SourceWriter::PushGenericNamespace(
     int modifiers) {
   GenericNamespace* generic_namespace;
diff --git a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.h b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.h
index 097887083e7..26b97f7a9c4 100644
--- a/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.h
+++ b/tensorflow-core/tensorflow-core-api/src/bazel/op_generator/source_writer.h
@@ -213,6 +213,7 @@ class SourceWriter {
   SourceWriter& WriteJavadoc(const Javadoc& javadoc);
   SourceWriter& WriteAnnotations(const std::list<Annotation>& annotations);
   SourceWriter& WriteGenerics(const std::list<const Type*>& generics);
+  SourceWriter& WriteTypeBounds(const std::list<Type>& bounds);
   GenericNamespace* PushGenericNamespace(int modifiers);
   void PopGenericNamespace();
 };
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataExperimentalOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataExperimentalOps.java
index cccc4ac8dcb..411d8e7969a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataExperimentalOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataExperimentalOps.java
@@ -18,12 +18,12 @@
 package org.tensorflow.op;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.data.experimental.DataServiceDataset;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 /**
  * An API for building {@code data.experimental} operations as {@link Op Op}s
@@ -54,7 +54,7 @@ public final class DataExperimentalOps {
   public DataServiceDataset dataServiceDataset(Operand<TInt64> datasetId,
       Operand<TString> processingMode, Operand<TString> address, Operand<TString> protocol,
       Operand<TString> jobName, Operand<TInt64> maxOutstandingRequests, Operand<?> iterationCounter,
-      List<DataType<?>> outputTypes, List<Shape> outputShapes,
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes,
       DataServiceDataset.Options... options) {
     return DataServiceDataset.create(scope, datasetId, processingMode, address, protocol, jobName, maxOutstandingRequests, iterationCounter, outputTypes, outputShapes, options);
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataOps.java
index 273025ef6bd..d01c9ee6d01 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DataOps.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.data.AnonymousIterator;
@@ -49,6 +48,7 @@
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 /**
  * An API for building {@code data} operations as {@link Op Op}s
@@ -72,7 +72,7 @@ public final class DataOps {
    * @param outputShapes
    * @return a new instance of AnonymousIterator
    */
-  public AnonymousIterator anonymousIterator(List<DataType<?>> outputTypes,
+  public AnonymousIterator anonymousIterator(List<Class<? extends TType>> outputTypes,
       List<Shape> outputShapes) {
     return AnonymousIterator.create(scope, outputTypes, outputShapes);
   }
@@ -90,8 +90,8 @@ public AnonymousIterator anonymousIterator(List<DataType<?>> outputTypes,
    * @return a new instance of BatchDataset
    */
   public BatchDataset batchDataset(Operand<?> inputDataset, Operand<TInt64> batchSize,
-      Operand<TBool> dropRemainder, List<DataType<?>> outputTypes, List<Shape> outputShapes,
-      BatchDataset.Options... options) {
+      Operand<TBool> dropRemainder, List<Class<? extends TType>> outputTypes,
+      List<Shape> outputShapes, BatchDataset.Options... options) {
     return BatchDataset.create(scope, inputDataset, batchSize, dropRemainder, outputTypes, outputShapes, options);
   }
 
@@ -126,7 +126,7 @@ public CSVDataset cSVDataset(Operand<TString> filenames, Operand<TString> compre
    * @return a new instance of ConcatenateDataset
    */
   public ConcatenateDataset concatenateDataset(Operand<?> inputDataset, Operand<?> anotherDataset,
-      List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return ConcatenateDataset.create(scope, inputDataset, anotherDataset, outputTypes, outputShapes);
   }
 
@@ -161,8 +161,8 @@ public DeserializeIterator deserializeIterator(Operand<?> resourceHandle, Operan
    * @param outputShapes
    * @return a new instance of Iterator
    */
-  public Iterator iterator(String sharedName, String container, List<DataType<?>> outputTypes,
-      List<Shape> outputShapes) {
+  public Iterator iterator(String sharedName, String container,
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return Iterator.create(scope, sharedName, container, outputTypes, outputShapes);
   }
 
@@ -174,8 +174,8 @@ public Iterator iterator(String sharedName, String container, List<DataType<?>>
    * @param outputShapes
    * @return a new instance of IteratorGetNext
    */
-  public IteratorGetNext iteratorGetNext(Operand<?> iterator, List<DataType<?>> outputTypes,
-      List<Shape> outputShapes) {
+  public IteratorGetNext iteratorGetNext(Operand<?> iterator,
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return IteratorGetNext.create(scope, iterator, outputTypes, outputShapes);
   }
 
@@ -188,7 +188,7 @@ public IteratorGetNext iteratorGetNext(Operand<?> iterator, List<DataType<?>> ou
    * @return a new instance of IteratorGetNextAsOptional
    */
   public IteratorGetNextAsOptional iteratorGetNextAsOptional(Operand<?> iterator,
-      List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return IteratorGetNextAsOptional.create(scope, iterator, outputTypes, outputShapes);
   }
 
@@ -205,8 +205,8 @@ public IteratorGetNextAsOptional iteratorGetNextAsOptional(Operand<?> iterator,
    * @param outputShapes
    * @return a new instance of IteratorGetNextSync
    */
-  public IteratorGetNextSync iteratorGetNextSync(Operand<?> iterator, List<DataType<?>> outputTypes,
-      List<Shape> outputShapes) {
+  public IteratorGetNextSync iteratorGetNextSync(Operand<?> iterator,
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return IteratorGetNextSync.create(scope, iterator, outputTypes, outputShapes);
   }
 
@@ -252,8 +252,8 @@ public OptionalFromValue optionalFromValue(Iterable<Operand<?>> components) {
    * @param outputShapes
    * @return a new instance of OptionalGetValue
    */
-  public OptionalGetValue optionalGetValue(Operand<?> optional, List<DataType<?>> outputTypes,
-      List<Shape> outputShapes) {
+  public OptionalGetValue optionalGetValue(Operand<?> optional,
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return OptionalGetValue.create(scope, optional, outputTypes, outputShapes);
   }
 
@@ -287,7 +287,7 @@ public OptionalNone optionalNone() {
    * @return a new instance of RangeDataset
    */
   public RangeDataset rangeDataset(Operand<TInt64> start, Operand<TInt64> stop,
-      Operand<TInt64> step, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      Operand<TInt64> step, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return RangeDataset.create(scope, start, stop, step, outputTypes, outputShapes);
   }
 
@@ -302,7 +302,7 @@ public RangeDataset rangeDataset(Operand<TInt64> start, Operand<TInt64> stop,
    * @return a new instance of RepeatDataset
    */
   public RepeatDataset repeatDataset(Operand<?> inputDataset, Operand<TInt64> count,
-      List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return RepeatDataset.create(scope, inputDataset, count, outputTypes, outputShapes);
   }
 
@@ -329,7 +329,7 @@ public SerializeIterator serializeIterator(Operand<?> resourceHandle,
    * @return a new instance of SkipDataset
    */
   public SkipDataset skipDataset(Operand<?> inputDataset, Operand<TInt64> count,
-      List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return SkipDataset.create(scope, inputDataset, count, outputTypes, outputShapes);
   }
 
@@ -345,7 +345,7 @@ public SkipDataset skipDataset(Operand<?> inputDataset, Operand<TInt64> count,
    * @return a new instance of TakeDataset
    */
   public TakeDataset takeDataset(Operand<?> inputDataset, Operand<TInt64> count,
-      List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return TakeDataset.create(scope, inputDataset, count, outputTypes, outputShapes);
   }
 
@@ -406,8 +406,8 @@ public TfRecordDataset tfRecordDataset(Operand<TString> filenames,
    * @param outputShapes
    * @return a new instance of ZipDataset
    */
-  public ZipDataset zipDataset(Iterable<Operand<?>> inputDatasets, List<DataType<?>> outputTypes,
-      List<Shape> outputShapes) {
+  public ZipDataset zipDataset(Iterable<Operand<?>> inputDatasets,
+      List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     return ZipDataset.create(scope, inputDatasets, outputTypes, outputShapes);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DebuggingOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DebuggingOps.java
deleted file mode 100644
index f12d18f925b..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DebuggingOps.java
+++ /dev/null
@@ -1,50 +0,0 @@
-// Copyright 2020 The TensorFlow Authors. All Rights Reserved.
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-// ==============================================================================
-//
-// This class has been generated, DO NOT EDIT!
-//
-package org.tensorflow.op;
-
-import org.tensorflow.Operand;
-import org.tensorflow.op.debugging.CheckNumerics;
-import org.tensorflow.types.family.TNumber;
-
-/**
- * An API for building {@code debugging} operations as {@link Op Op}s
- *
- * @see {@link Ops}
- */
-public final class DebuggingOps {
-  private final Scope scope;
-
-  DebuggingOps(Scope scope) {
-    this.scope = scope;
-  }
-
-  /**
-   * Checks a tensor for NaN and Inf values.
-   *  <p>
-   *  When run, reports an `InvalidArgument` error if `tensor` has any values
-   *  that are not a number (NaN) or infinity (Inf). Otherwise, passes `tensor` as-is.
-   *
-   * @param <T> data type for {@code output()} output
-   * @param tensor
-   * @param message Prefix of the error message.
-   * @return a new instance of CheckNumerics
-   */
-  public <T extends TNumber> CheckNumerics<T> checkNumerics(Operand<T> tensor, String message) {
-    return CheckNumerics.create(scope, tensor, message);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DtypesOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DtypesOps.java
index 16d571a6428..824f3383f2b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DtypesOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/DtypesOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.dtypes.AsString;
 import org.tensorflow.op.dtypes.Cast;
@@ -70,7 +69,7 @@ public <T extends TType> AsString asString(Operand<T> input, AsString.Options...
    * @param options carries optional attributes values
    * @return a new instance of Cast
    */
-  public <U extends TType, T extends TType> Cast<U> cast(Operand<T> x, DataType<U> DstT,
+  public <U extends TType, T extends TType> Cast<U> cast(Operand<T> x, Class<U> DstT,
       Cast.Options... options) {
     return Cast.create(scope, x, DstT, options);
   }
@@ -99,7 +98,7 @@ public <U extends TType, T extends TType> Cast<U> cast(Operand<T> x, DataType<U>
    * @return a new instance of Complex
    */
   public <U extends TType, T extends TNumber> Complex<U> complex(Operand<T> real, Operand<T> imag,
-      DataType<U> Tout) {
+      Class<U> Tout) {
     return Complex.create(scope, real, imag, Tout);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ImageOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ImageOps.java
index eea2fc4b8f1..b0fa7751b8a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ImageOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ImageOps.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.image.AdjustContrast;
 import org.tensorflow.op.image.AdjustHue;
@@ -269,7 +268,7 @@ public <T extends TNumber> CropAndResizeGradBoxes cropAndResizeGradBoxes(Operand
    */
   public <T extends TNumber> CropAndResizeGradImage<T> cropAndResizeGradImage(
       Operand<TFloat32> grads, Operand<TFloat32> boxes, Operand<TInt32> boxInd,
-      Operand<TInt32> imageSize, DataType<T> T, CropAndResizeGradImage.Options... options) {
+      Operand<TInt32> imageSize, Class<T> T, CropAndResizeGradImage.Options... options) {
     return CropAndResizeGradImage.create(scope, grads, boxes, boxInd, imageSize, T, options);
   }
 
@@ -460,7 +459,7 @@ public DecodePng<TUint8> decodePng(Operand<TString> contents, DecodePng.Options.
    * @param options carries optional attributes values
    * @return a new instance of DecodePng
    */
-  public <T extends TNumber> DecodePng<T> decodePng(Operand<TString> contents, DataType<T> dtype,
+  public <T extends TNumber> DecodePng<T> decodePng(Operand<TString> contents, Class<T> dtype,
       DecodePng.Options... options) {
     return DecodePng.create(scope, contents, dtype, options);
   }
@@ -622,7 +621,7 @@ public ExtractJpegShape<TInt32> extractJpegShape(Operand<TString> contents) {
    * @return a new instance of ExtractJpegShape
    */
   public <T extends TNumber> ExtractJpegShape<T> extractJpegShape(Operand<TString> contents,
-      DataType<T> outputType) {
+      Class<T> outputType) {
     return ExtractJpegShape.create(scope, contents, outputType);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/IoOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/IoOps.java
index adc656dc5af..78aa5897a97 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/IoOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/IoOps.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.io.DecodeBase64;
@@ -168,7 +167,7 @@ public DecodeJsonExample decodeJsonExample(Operand<TString> jsonExamples) {
    * @return a new instance of DecodePaddedRaw
    */
   public <T extends TNumber> DecodePaddedRaw<T> decodePaddedRaw(Operand<TString> inputBytes,
-      Operand<TInt32> fixedLength, DataType<T> outType, DecodePaddedRaw.Options... options) {
+      Operand<TInt32> fixedLength, Class<T> outType, DecodePaddedRaw.Options... options) {
     return DecodePaddedRaw.create(scope, inputBytes, fixedLength, outType, options);
   }
 
@@ -181,7 +180,7 @@ public <T extends TNumber> DecodePaddedRaw<T> decodePaddedRaw(Operand<TString> i
    * @param options carries optional attributes values
    * @return a new instance of DecodeRaw
    */
-  public <T extends TType> DecodeRaw<T> decodeRaw(Operand<TString> bytes, DataType<T> outType,
+  public <T extends TType> DecodeRaw<T> decodeRaw(Operand<TString> bytes, Class<T> outType,
       DecodeRaw.Options... options) {
     return DecodeRaw.create(scope, bytes, outType, options);
   }
@@ -238,7 +237,7 @@ public <T extends TType> DecodeRaw<T> decodeRaw(Operand<TString> bytes, DataType
    * @return a new instance of DeserializeManySparse
    */
   public <T extends TType> DeserializeManySparse<T> deserializeManySparse(
-      Operand<TString> serializedSparse, DataType<T> dtype) {
+      Operand<TString> serializedSparse, Class<T> dtype) {
     return DeserializeManySparse.create(scope, serializedSparse, dtype);
   }
 
@@ -267,7 +266,8 @@ public EncodeBase64 encodeBase64(Operand<TString> input, EncodeBase64.Options...
    * @param options carries optional attributes values
    * @return a new instance of FifoQueue
    */
-  public FifoQueue fifoQueue(List<DataType<?>> componentTypes, FifoQueue.Options... options) {
+  public FifoQueue fifoQueue(List<Class<? extends TType>> componentTypes,
+      FifoQueue.Options... options) {
     return FifoQueue.create(scope, componentTypes, options);
   }
 
@@ -331,7 +331,7 @@ public MatchingFiles matchingFiles(Operand<TString> pattern) {
    * @param options carries optional attributes values
    * @return a new instance of PaddingFifoQueue
    */
-  public PaddingFifoQueue paddingFifoQueue(List<DataType<?>> componentTypes,
+  public PaddingFifoQueue paddingFifoQueue(List<Class<? extends TType>> componentTypes,
       PaddingFifoQueue.Options... options) {
     return PaddingFifoQueue.create(scope, componentTypes, options);
   }
@@ -393,9 +393,9 @@ public PaddingFifoQueue paddingFifoQueue(List<DataType<?>> componentTypes,
    */
   public ParseExample parseExample(Operand<TString> serialized, Operand<TString> names,
       Operand<TString> sparseKeys, Operand<TString> denseKeys, Operand<TString> raggedKeys,
-      Iterable<Operand<?>> denseDefaults, Long numSparse, List<DataType<?>> sparseTypes,
-      List<DataType<?>> raggedValueTypes, List<DataType<?>> raggedSplitTypes,
-      List<Shape> denseShapes) {
+      Iterable<Operand<?>> denseDefaults, Long numSparse, List<Class<? extends TType>> sparseTypes,
+      List<Class<? extends TType>> raggedValueTypes,
+      List<Class<? extends TNumber>> raggedSplitTypes, List<Shape> denseShapes) {
     return ParseExample.create(scope, serialized, names, sparseKeys, denseKeys, raggedKeys, denseDefaults, numSparse, sparseTypes, raggedValueTypes, raggedSplitTypes, denseShapes);
   }
 
@@ -451,10 +451,13 @@ public ParseSequenceExample parseSequenceExample(Operand<TString> serialized,
       Operand<TString> contextDenseKeys, Operand<TString> contextRaggedKeys,
       Operand<TString> featureListSparseKeys, Operand<TString> featureListDenseKeys,
       Operand<TString> featureListRaggedKeys, Operand<TBool> featureListDenseMissingAssumedEmpty,
-      Iterable<Operand<?>> contextDenseDefaults, List<DataType<?>> contextSparseTypes,
-      List<DataType<?>> contextRaggedValueTypes, List<DataType<?>> contextRaggedSplitTypes,
-      List<DataType<?>> featureListDenseTypes, List<DataType<?>> featureListSparseTypes,
-      List<DataType<?>> featureListRaggedValueTypes, List<DataType<?>> featureListRaggedSplitTypes,
+      Iterable<Operand<?>> contextDenseDefaults, List<Class<? extends TType>> contextSparseTypes,
+      List<Class<? extends TType>> contextRaggedValueTypes,
+      List<Class<? extends TNumber>> contextRaggedSplitTypes,
+      List<Class<? extends TType>> featureListDenseTypes,
+      List<Class<? extends TType>> featureListSparseTypes,
+      List<Class<? extends TType>> featureListRaggedValueTypes,
+      List<Class<? extends TNumber>> featureListRaggedSplitTypes,
       ParseSequenceExample.Options... options) {
     return ParseSequenceExample.create(scope, serialized, debugName, contextSparseKeys, contextDenseKeys, contextRaggedKeys, featureListSparseKeys, featureListDenseKeys, featureListRaggedKeys, featureListDenseMissingAssumedEmpty, contextDenseDefaults, contextSparseTypes, contextRaggedValueTypes, contextRaggedSplitTypes, featureListDenseTypes, featureListSparseTypes, featureListRaggedValueTypes, featureListRaggedSplitTypes, options);
   }
@@ -496,7 +499,7 @@ public ParseSequenceExample parseSequenceExample(Operand<TString> serialized,
    */
   public ParseSingleExample parseSingleExample(Operand<TString> serialized,
       Iterable<Operand<?>> denseDefaults, Long numSparse, List<String> sparseKeys,
-      List<String> denseKeys, List<DataType<?>> sparseTypes, List<Shape> denseShapes) {
+      List<String> denseKeys, List<Class<? extends TType>> sparseTypes, List<Shape> denseShapes) {
     return ParseSingleExample.create(scope, serialized, denseDefaults, numSparse, sparseKeys, denseKeys, sparseTypes, denseShapes);
   }
 
@@ -550,8 +553,9 @@ public ParseSingleSequenceExample parseSingleSequenceExample(Operand<TString> se
       Iterable<Operand<TString>> contextSparseKeys, Iterable<Operand<TString>> contextDenseKeys,
       Iterable<Operand<TString>> featureListSparseKeys,
       Iterable<Operand<TString>> featureListDenseKeys, Iterable<Operand<?>> contextDenseDefaults,
-      Operand<TString> debugName, List<DataType<?>> contextSparseTypes,
-      List<DataType<?>> featureListDenseTypes, List<DataType<?>> featureListSparseTypes,
+      Operand<TString> debugName, List<Class<? extends TType>> contextSparseTypes,
+      List<Class<? extends TType>> featureListDenseTypes,
+      List<Class<? extends TType>> featureListSparseTypes,
       ParseSingleSequenceExample.Options... options) {
     return ParseSingleSequenceExample.create(scope, serialized, featureListDenseMissingAssumedEmpty, contextSparseKeys, contextDenseKeys, featureListSparseKeys, featureListDenseKeys, contextDenseDefaults, debugName, contextSparseTypes, featureListDenseTypes, featureListSparseTypes, options);
   }
@@ -566,7 +570,7 @@ public ParseSingleSequenceExample parseSingleSequenceExample(Operand<TString> se
    * @return a new instance of ParseTensor
    */
   public <T extends TType> ParseTensor<T> parseTensor(Operand<TString> serialized,
-      DataType<T> outType) {
+      Class<T> outType) {
     return ParseTensor.create(scope, serialized, outType);
   }
 
@@ -587,8 +591,8 @@ public <T extends TType> ParseTensor<T> parseTensor(Operand<TString> serialized,
    * @param options carries optional attributes values
    * @return a new instance of PriorityQueue
    */
-  public PriorityQueue priorityQueue(List<DataType<?>> componentTypes, List<Shape> shapes,
-      PriorityQueue.Options... options) {
+  public PriorityQueue priorityQueue(List<Class<? extends TType>> componentTypes,
+      List<Shape> shapes, PriorityQueue.Options... options) {
     return PriorityQueue.create(scope, componentTypes, shapes, options);
   }
 
@@ -624,7 +628,7 @@ public QueueClose queueClose(Operand<?> handle, QueueClose.Options... options) {
    * @param options carries optional attributes values
    * @return a new instance of QueueDequeue
    */
-  public QueueDequeue queueDequeue(Operand<?> handle, List<DataType<?>> componentTypes,
+  public QueueDequeue queueDequeue(Operand<?> handle, List<Class<? extends TType>> componentTypes,
       QueueDequeue.Options... options) {
     return QueueDequeue.create(scope, handle, componentTypes, options);
   }
@@ -653,7 +657,7 @@ public QueueDequeue queueDequeue(Operand<?> handle, List<DataType<?>> componentT
    * @return a new instance of QueueDequeueMany
    */
   public QueueDequeueMany queueDequeueMany(Operand<?> handle, Operand<TInt32> n,
-      List<DataType<?>> componentTypes, QueueDequeueMany.Options... options) {
+      List<Class<? extends TType>> componentTypes, QueueDequeueMany.Options... options) {
     return QueueDequeueMany.create(scope, handle, n, componentTypes, options);
   }
 
@@ -685,7 +689,7 @@ public QueueDequeueMany queueDequeueMany(Operand<?> handle, Operand<TInt32> n,
    * @return a new instance of QueueDequeueUpTo
    */
   public QueueDequeueUpTo queueDequeueUpTo(Operand<?> handle, Operand<TInt32> n,
-      List<DataType<?>> componentTypes, QueueDequeueUpTo.Options... options) {
+      List<Class<? extends TType>> componentTypes, QueueDequeueUpTo.Options... options) {
     return QueueDequeueUpTo.create(scope, handle, n, componentTypes, options);
   }
 
@@ -762,7 +766,7 @@ public QueueSize queueSize(Operand<?> handle) {
    * @param options carries optional attributes values
    * @return a new instance of RandomShuffleQueue
    */
-  public RandomShuffleQueue randomShuffleQueue(List<DataType<?>> componentTypes,
+  public RandomShuffleQueue randomShuffleQueue(List<Class<? extends TType>> componentTypes,
       RandomShuffleQueue.Options... options) {
     return RandomShuffleQueue.create(scope, componentTypes, options);
   }
@@ -914,7 +918,7 @@ public <T extends TType> SerializeManySparse<TString> serializeManySparse(
    */
   public <U extends TType, T extends TType> SerializeManySparse<U> serializeManySparse(
       Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape,
-      DataType<U> outType) {
+      Class<U> outType) {
     return SerializeManySparse.create(scope, sparseIndices, sparseValues, sparseShape, outType);
   }
 
@@ -945,7 +949,7 @@ public <T extends TType> SerializeSparse<TString> serializeSparse(Operand<TInt64
    */
   public <U extends TType, T extends TType> SerializeSparse<U> serializeSparse(
       Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape,
-      DataType<U> outType) {
+      Class<U> outType) {
     return SerializeSparse.create(scope, sparseIndices, sparseValues, sparseShape, outType);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/LinalgOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/LinalgOps.java
index b2242f1068e..8163e0a50b3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/LinalgOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/LinalgOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.linalg.BandPart;
 import org.tensorflow.op.linalg.BatchCholesky;
@@ -396,7 +395,7 @@ public <T extends TType> Det<T> det(Operand<T> input) {
    * @param options carries optional attributes values
    * @return a new instance of Eig
    */
-  public <U extends TType, T extends TType> Eig<U> eig(Operand<T> input, DataType<U> Tout,
+  public <U extends TType, T extends TType> Eig<U> eig(Operand<T> input, Class<U> Tout,
       Eig.Options... options) {
     return Eig.create(scope, input, Tout, options);
   }
@@ -682,7 +681,7 @@ public <T extends TType> Lu<T, TInt32> lu(Operand<T> input) {
    * @return a new instance of Lu
    */
   public <T extends TType, U extends TNumber> Lu<T, U> lu(Operand<T> input,
-      DataType<U> outputIdxType) {
+      Class<U> outputIdxType) {
     return Lu.create(scope, input, outputIdxType);
   }
 
@@ -1373,7 +1372,7 @@ public <T extends TType> Qr<T> qr(Operand<T> input, Qr.Options... options) {
    */
   public <V extends TType, T extends TType, U extends TType, W extends TType> QuantizedMatMul<V> quantizedMatMul(
       Operand<T> a, Operand<U> b, Operand<TFloat32> minA, Operand<TFloat32> maxA,
-      Operand<TFloat32> minB, Operand<TFloat32> maxB, DataType<V> Toutput, DataType<W> Tactivation,
+      Operand<TFloat32> minB, Operand<TFloat32> maxB, Class<V> Toutput, Class<W> Tactivation,
       QuantizedMatMul.Options... options) {
     return QuantizedMatMul.create(scope, a, b, minA, maxA, minB, maxB, Toutput, Tactivation, options);
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/LinalgSparseOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/LinalgSparseOps.java
deleted file mode 100644
index 7f8777c883a..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/LinalgSparseOps.java
+++ /dev/null
@@ -1,463 +0,0 @@
-// Copyright 2020 The TensorFlow Authors. All Rights Reserved.
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-// ==============================================================================
-//
-// This class has been generated, DO NOT EDIT!
-//
-package org.tensorflow.op;
-
-import org.tensorflow.DataType;
-import org.tensorflow.Operand;
-import org.tensorflow.op.linalg.sparse.CSRSparseMatrixToSparseTensor;
-import org.tensorflow.op.linalg.sparse.DenseToCSRSparseMatrix;
-import org.tensorflow.op.linalg.sparse.SparseMatrixAdd;
-import org.tensorflow.op.linalg.sparse.SparseMatrixMatMul;
-import org.tensorflow.op.linalg.sparse.SparseMatrixMul;
-import org.tensorflow.op.linalg.sparse.SparseMatrixNNZ;
-import org.tensorflow.op.linalg.sparse.SparseMatrixOrderingAMD;
-import org.tensorflow.op.linalg.sparse.SparseMatrixSoftmax;
-import org.tensorflow.op.linalg.sparse.SparseMatrixSoftmaxGrad;
-import org.tensorflow.op.linalg.sparse.SparseMatrixSparseCholesky;
-import org.tensorflow.op.linalg.sparse.SparseMatrixSparseMatMul;
-import org.tensorflow.op.linalg.sparse.SparseMatrixTranspose;
-import org.tensorflow.op.linalg.sparse.SparseMatrixZeros;
-import org.tensorflow.op.linalg.sparse.SparseTensorToCSRSparseMatrix;
-import org.tensorflow.types.TInt32;
-import org.tensorflow.types.TInt64;
-import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
-
-/**
- * An API for building {@code linalg.sparse} operations as {@link Op Op}s
- *
- * @see {@link Ops}
- */
-public final class LinalgSparseOps {
-  private final Scope scope;
-
-  LinalgSparseOps(Scope scope) {
-    this.scope = scope;
-  }
-
-  /**
-   * Converts a (possibly batched) CSRSparesMatrix to a SparseTensor.
-   *
-   * @param <T> data type for {@code values()} output
-   * @param sparseMatrix A (possibly batched) CSRSparseMatrix.
-   * @param type
-   * @return a new instance of CSRSparseMatrixToSparseTensor
-   */
-  public <T extends TType> CSRSparseMatrixToSparseTensor<T> cSRSparseMatrixToSparseTensor(
-      Operand<?> sparseMatrix, DataType<T> type) {
-    return CSRSparseMatrixToSparseTensor.create(scope, sparseMatrix, type);
-  }
-
-  /**
-   * Converts a dense tensor to a (possibly batched) CSRSparseMatrix.
-   *
-   * @param denseInput A Dense tensor.
-   * @param indices Indices of nonzero elements.
-   * @return a new instance of DenseToCSRSparseMatrix
-   */
-  public <T extends TType> DenseToCSRSparseMatrix denseToCSRSparseMatrix(Operand<T> denseInput,
-      Operand<TInt64> indices) {
-    return DenseToCSRSparseMatrix.create(scope, denseInput, indices);
-  }
-
-  /**
-   * Sparse addition of two CSR matrices, C = alpha * A + beta * B.
-   *  <p>
-   *  The gradients of SparseMatrixAdd outputs with respect to alpha and beta are not
-   *  currently defined (TensorFlow will return zeros for these entries).
-   *
-   * @param a A CSRSparseMatrix.
-   * @param b A CSRSparseMatrix.
-   * @param alpha A constant scalar.
-   * @param beta A constant scalar.
-   * @return a new instance of SparseMatrixAdd
-   */
-  public <T extends TType> SparseMatrixAdd sparseMatrixAdd(Operand<?> a, Operand<?> b,
-      Operand<T> alpha, Operand<T> beta) {
-    return SparseMatrixAdd.create(scope, a, b, alpha, beta);
-  }
-
-  /**
-   * Matrix-multiplies a sparse matrix with a dense matrix.
-   *  <p>
-   *  Returns a dense matrix.
-   *  For inputs A and B, where A is CSR and B is dense; this op returns a dense C;
-   *  <p>
-   *  If transpose_output is false, returns:
-   *  <pre>{@code
-   *    C = A . B
-   *  }</pre>
-   *  If transpose_output is `true`, returns:
-   *  <pre>{@code
-   *    C = transpose(A . B) = transpose(B) . transpose(A)
-   *  }</pre>
-   *  where the transposition is performed along the two innermost (matrix)
-   *  dimensions.
-   *  <p>
-   *  If conjugate_output is `true`, returns:
-   *  <pre>{@code
-   *    C = conjugate(A . B) = conjugate(A) . conjugate(B)
-   *  }</pre>
-   *  If both conjugate_output and transpose_output are `true`, returns:
-   *  <pre>{@code
-   *    C = conjugate(transpose(A . B)) = conjugate(transpose(B)) .
-   *                                      conjugate(transpose(A))
-   *  }</pre>
-   *
-   * @param <T> data type for {@code output()} output
-   * @param a A CSRSparseMatrix.
-   * @param b A dense tensor.
-   * @param options carries optional attributes values
-   * @return a new instance of SparseMatrixMatMul
-   */
-  public <T extends TType> SparseMatrixMatMul<T> sparseMatrixMatMul(Operand<?> a, Operand<T> b,
-      SparseMatrixMatMul.Options... options) {
-    return SparseMatrixMatMul.create(scope, a, b, options);
-  }
-
-  /**
-   * Element-wise multiplication of a sparse matrix with a dense tensor.
-   *  <p>
-   *  Returns a sparse matrix.
-   *  <p>
-   *  The dense tensor `b` may be either a scalar; otherwise `a` must be a rank-3
-   *  `SparseMatrix`; in this case `b` must be shaped `[batch_size, 1, 1]` and the
-   *  multiply operation broadcasts.
-   *  <p>
-   *  <b>NOTE</b> even if `b` is zero, the sparsity structure of the output does not
-   *  change.
-   *
-   * @param a A CSRSparseMatrix.
-   * @param b A dense tensor.
-   * @return a new instance of SparseMatrixMul
-   */
-  public <T extends TType> SparseMatrixMul sparseMatrixMul(Operand<?> a, Operand<T> b) {
-    return SparseMatrixMul.create(scope, a, b);
-  }
-
-  /**
-   * Returns the number of nonzeroes of `sparse_matrix`.
-   *
-   * @param sparseMatrix A CSRSparseMatrix.
-   * @return a new instance of SparseMatrixNNZ
-   */
-  public SparseMatrixNNZ sparseMatrixNNZ(Operand<?> sparseMatrix) {
-    return SparseMatrixNNZ.create(scope, sparseMatrix);
-  }
-
-  /**
-   * Computes the Approximate Minimum Degree (AMD) ordering of `input`.
-   *  <p>
-   *  Computes the Approximate Minimum Degree (AMD) ordering for a sparse matrix.
-   *  <p>
-   *  The returned permutation may be used to permute the rows and columns of the
-   *  given sparse matrix. This typically results in permuted sparse matrix's sparse
-   *  Cholesky (or other decompositions) in having fewer zero fill-in compared to
-   *  decomposition of the original matrix.
-   *  <p>
-   *  The input sparse matrix may have rank 2 or rank 3. The output Tensor,
-   *  representing would then have rank 1 or 2 respectively, with the same batch
-   *  shape as the input.
-   *  <p>
-   *  Each component of the input sparse matrix must represent a square symmetric
-   *  matrix; only the lower triangular part of the matrix is read. The values of the
-   *  sparse matrix does not affect the returned permutation, only the sparsity
-   *  pattern of the sparse matrix is used. Hence, a single AMD ordering may be
-   *  reused for the Cholesky decompositions of sparse matrices with the same sparsity
-   *  pattern but with possibly different values.
-   *  <p>
-   *  Each batch component of the output permutation represents a permutation of `N`
-   *  elements, where the input sparse matrix components each have `N` rows. That is,
-   *  the component contains each of the integers `{0, .. N-1}` exactly once. The
-   *  `i`th element represents the row index that the `i`th row maps to.
-   *  <p>
-   *  Usage example:
-   *  <pre>{@code
-   *      from tensorflow.python.ops.linalg.sparse import sparse_csr_matrix_ops
-   *
-   *      a_indices = np.array([[0, 0], [1, 1], [2, 1], [2, 2], [3, 3]])
-   *      a_values = np.array([1.0, 2.0, 1.0, 3.0, 4.0], np.float32)
-   *      a_dense_shape = [4, 4]
-   *
-   *      with tf.Session() as sess:
-   *        # Define (COO format) SparseTensor over Numpy array.
-   *        a_st = tf.SparseTensor(a_indices, a_values, a_dense_shape)
-   *
-   *        # Convert SparseTensors to CSR SparseMatrix.
-   *        a_sm = sparse_csr_matrix_ops.sparse_tensor_to_csr_sparse_matrix(
-   *            a_st.indices, a_st.values, a_st.dense_shape)
-   *
-   *        # Obtain the AMD Ordering for the CSR SparseMatrix.
-   *        ordering_amd = sparse_csr_matrix_ops.sparse_matrix_ordering_amd(sparse_matrix)
-   *
-   *        ordering_amd_value = sess.run(ordering_amd)
-   *  }</pre>
-   *  `ordering_amd_value` stores the AMD ordering: `[1 2 3 0]`.
-   *  <p>
-   *  input: A `CSRSparseMatrix`.
-   *
-   * @param input A `CSRSparseMatrix`.
-   * @return a new instance of SparseMatrixOrderingAMD
-   */
-  public SparseMatrixOrderingAMD sparseMatrixOrderingAMD(Operand<?> input) {
-    return SparseMatrixOrderingAMD.create(scope, input);
-  }
-
-  /**
-   * Calculates the softmax of a CSRSparseMatrix.
-   *  <p>
-   *  Calculate the softmax of the innermost dimensions of a SparseMatrix.
-   *  <p>
-   *  Missing values are treated as `-inf` (i.e., logits of zero probability); and
-   *  the output has the same sparsity structure as the input (though missing values
-   *  in the output may now be treated as having probability zero).
-   *
-   * @param logits A CSRSparseMatrix.
-   * @param type
-   * @return a new instance of SparseMatrixSoftmax
-   */
-  public <T extends TNumber> SparseMatrixSoftmax sparseMatrixSoftmax(Operand<?> logits,
-      DataType<T> type) {
-    return SparseMatrixSoftmax.create(scope, logits, type);
-  }
-
-  /**
-   * Calculates the gradient of the SparseMatrixSoftmax op.
-   *
-   * @param softmax A CSRSparseMatrix.
-   * @param gradSoftmax The gradient of `softmax`.
-   * @param type
-   * @return a new instance of SparseMatrixSoftmaxGrad
-   */
-  public <T extends TNumber> SparseMatrixSoftmaxGrad sparseMatrixSoftmaxGrad(Operand<?> softmax,
-      Operand<?> gradSoftmax, DataType<T> type) {
-    return SparseMatrixSoftmaxGrad.create(scope, softmax, gradSoftmax, type);
-  }
-
-  /**
-   * Computes the sparse Cholesky decomposition of `input`.
-   *  <p>
-   *  Computes the Sparse Cholesky decomposition of a sparse matrix, with the given
-   *  fill-in reducing permutation.
-   *  <p>
-   *  The input sparse matrix and the fill-in reducing permutation `permutation` must
-   *  have compatible shapes. If the sparse matrix has rank 3; with the batch
-   *  dimension `B`, then the `permutation` must be of rank 2; with the same batch
-   *  dimension `B`. There is no support for broadcasting.
-   *  <p>
-   *  Furthermore, each component vector of `permutation` must be of length `N`,
-   *  containing each of the integers {0, 1, ..., N - 1} exactly once, where `N` is
-   *  the number of rows of each component of the sparse matrix.
-   *  <p>
-   *  Each component of the input sparse matrix must represent a symmetric positive
-   *  definite (SPD) matrix; although only the lower triangular part of the matrix is
-   *  read. If any individual component is not SPD, then an InvalidArgument error is
-   *  thrown.
-   *  <p>
-   *  The returned sparse matrix has the same dense shape as the input sparse matrix.
-   *  For each component `A` of the input sparse matrix, the corresponding output
-   *  sparse matrix represents `L`, the lower triangular Cholesky factor satisfying
-   *  the following identity:
-   *  <pre>{@code
-   *    A = L * Lt
-   *  }</pre>
-   *  where Lt denotes the transpose of L (or its conjugate transpose, if `type` is
-   *  `complex64` or `complex128`).
-   *  <p>
-   *  The `type` parameter denotes the type of the matrix elements. The supported
-   *  types are: `float32`, `float64`, `complex64` and `complex128`.
-   *  <p>
-   *  Usage example:
-   *  <pre>{@code
-   *      from tensorflow.python.ops.linalg.sparse import sparse_csr_matrix_ops
-   *
-   *      a_indices = np.array([[0, 0], [1, 1], [2, 1], [2, 2], [3, 3]])
-   *      a_values = np.array([1.0, 2.0, 1.0, 3.0, 4.0], np.float32)
-   *      a_dense_shape = [4, 4]
-   *
-   *      with tf.Session() as sess:
-   *        # Define (COO format) SparseTensor over Numpy array.
-   *        a_st = tf.SparseTensor(a_indices, a_values, a_dense_shape)
-   *
-   *        # Convert SparseTensors to CSR SparseMatrix.
-   *        a_sm = sparse_csr_matrix_ops.sparse_tensor_to_csr_sparse_matrix(
-   *            a_st.indices, a_st.values, a_st.dense_shape)
-   *
-   *        # Obtain the Sparse Cholesky factor using AMD Ordering for reducing zero
-   *        # fill-in (number of structural non-zeros in the sparse Cholesky factor).
-   *        ordering_amd = sparse_csr_matrix_ops.sparse_matrix_ordering_amd(sparse_matrix)
-   *        cholesky_sparse_matrices = (
-   *            sparse_csr_matrix_ops.sparse_matrix_sparse_cholesky(
-   *                sparse_matrix, ordering_amd, type=tf.float32))
-   *
-   *        # Convert the CSRSparseMatrix Cholesky factor to a dense Tensor
-   *        dense_cholesky = sparse_csr_matrix_ops.csr_sparse_matrix_to_dense(
-   *            cholesky_sparse_matrices, tf.float32)
-   *
-   *        # Evaluate the dense Tensor value.
-   *        dense_cholesky_value = sess.run(dense_cholesky)
-   *  }</pre>
-   *  `dense_cholesky_value` stores the dense Cholesky factor:
-   *  <pre>{@code
-   *      [[  1.  0.    0.    0.]
-   *       [  0.  1.41  0.    0.]
-   *       [  0.  0.70  1.58  0.]
-   *       [  0.  0.    0.    2.]]
-   *  }</pre>
-   *  input: A `CSRSparseMatrix`.
-   *  permutation: A `Tensor`.
-   *  type: The type of `input`.
-   *
-   * @param input A `CSRSparseMatrix`.
-   * @param permutation A fill-in reducing permutation matrix.
-   * @param type
-   * @return a new instance of SparseMatrixSparseCholesky
-   */
-  public <T extends TType> SparseMatrixSparseCholesky sparseMatrixSparseCholesky(Operand<?> input,
-      Operand<TInt32> permutation, DataType<T> type) {
-    return SparseMatrixSparseCholesky.create(scope, input, permutation, type);
-  }
-
-  /**
-   * Sparse-matrix-multiplies two CSR matrices `a` and `b`.
-   *  <p>
-   *  Performs a matrix multiplication of a sparse matrix `a` with a sparse matrix
-   *  `b`; returns a sparse matrix `a * b`, unless either `a` or `b` is transposed or
-   *  adjointed.
-   *  <p>
-   *  Each matrix may be transposed or adjointed (conjugated and transposed)
-   *  according to the Boolean parameters `transpose_a`, `adjoint_a`, `transpose_b`
-   *  and `adjoint_b`. At most one of `transpose_a` or `adjoint_a` may be True.
-   *  Similarly, at most one of `transpose_b` or `adjoint_b` may be True.
-   *  <p>
-   *  The inputs must have compatible shapes. That is, the inner dimension of `a`
-   *  must be equal to the outer dimension of `b`. This requirement is adjusted
-   *  according to whether either `a` or `b` is transposed or adjointed.
-   *  <p>
-   *  The `type` parameter denotes the type of the matrix elements. Both `a` and `b`
-   *  must have the same type. The supported types are: `float32`, `float64`,
-   *  `complex64` and `complex128`.
-   *  <p>
-   *  Both `a` and `b` must have the same rank. Broadcasting is not supported. If they
-   *  have rank 3, each batch of 2D CSRSparseMatrices within `a` and `b` must have the
-   *  same dense shape.
-   *  <p>
-   *  The sparse matrix product may have numeric (non-structural) zeros.
-   *  TODO(anudhyan): Consider adding a boolean attribute to control whether to prune
-   *  zeros.
-   *  <p>
-   *  Usage example:
-   *  <pre>{@code
-   *      from tensorflow.python.ops.linalg.sparse import sparse_csr_matrix_ops
-   *
-   *      a_indices = np.array([[0, 0], [2, 3], [2, 4], [3, 0]])
-   *      a_values = np.array([1.0, 5.0, -1.0, -2.0], np.float32)
-   *      a_dense_shape = [4, 5]
-   *
-   *      b_indices = np.array([[0, 0], [3, 0], [3, 1]])
-   *      b_values = np.array([2.0, 7.0, 8.0], np.float32)
-   *      b_dense_shape = [5, 3]
-   *
-   *      with tf.Session() as sess:
-   *        # Define (COO format) Sparse Tensors over Numpy arrays
-   *        a_st = tf.SparseTensor(a_indices, a_values, a_dense_shape)
-   *        b_st = tf.SparseTensor(b_indices, b_values, b_dense_shape)
-   *
-   *        # Convert SparseTensors to CSR SparseMatrix
-   *        a_sm = sparse_csr_matrix_ops.sparse_tensor_to_csr_sparse_matrix(
-   *            a_st.indices, a_st.values, a_st.dense_shape)
-   *        b_sm = sparse_csr_matrix_ops.sparse_tensor_to_csr_sparse_matrix(
-   *            b_st.indices, b_st.values, b_st.dense_shape)
-   *
-   *        # Compute the CSR SparseMatrix matrix multiplication
-   *        c_sm = sparse_csr_matrix_ops.sparse_matrix_sparse_mat_mul(
-   *            a=a_sm, b=b_sm, type=tf.float32)
-   *
-   *        # Convert the CSR SparseMatrix product to a dense Tensor
-   *        c_sm_dense = sparse_csr_matrix_ops.csr_sparse_matrix_to_dense(
-   *            c_sm, tf.float32)
-   *        # Evaluate the dense Tensor value
-   *        c_sm_dense_value = sess.run(c_sm_dense)
-   *  }</pre>
-   *  `c_sm_dense_value` stores the dense matrix product:
-   *  <pre>{@code
-   *      [[  2.   0.   0.]
-   *       [  0.   0.   0.]
-   *       [ 35.  40.   0.]
-   *       [ -4.   0.   0.]]
-   *  }</pre>
-   *  a: A `CSRSparseMatrix`.
-   *  b: A `CSRSparseMatrix` with the same type and rank as `a`.
-   *  type: The type of both `a` and `b`.
-   *  transpose_a: If True, `a` transposed before multiplication.
-   *  transpose_b: If True, `b` transposed before multiplication.
-   *  adjoint_a: If True, `a` adjointed before multiplication.
-   *  adjoint_b: If True, `b` adjointed before multiplication.
-   *
-   * @param a A CSRSparseMatrix.
-   * @param b A CSRSparseMatrix.
-   * @param type
-   * @param options carries optional attributes values
-   * @return a new instance of SparseMatrixSparseMatMul
-   */
-  public <T extends TType> SparseMatrixSparseMatMul sparseMatrixSparseMatMul(Operand<?> a,
-      Operand<?> b, DataType<T> type, SparseMatrixSparseMatMul.Options... options) {
-    return SparseMatrixSparseMatMul.create(scope, a, b, type, options);
-  }
-
-  /**
-   * Transposes the inner (matrix) dimensions of a CSRSparseMatrix.
-   *  <p>
-   *  Transposes the inner (matrix) dimensions of a SparseMatrix and optionally
-   *  conjugates its values.
-   *
-   * @param input A CSRSparseMatrix.
-   * @param type
-   * @param options carries optional attributes values
-   * @return a new instance of SparseMatrixTranspose
-   */
-  public <T extends TType> SparseMatrixTranspose sparseMatrixTranspose(Operand<?> input,
-      DataType<T> type, SparseMatrixTranspose.Options... options) {
-    return SparseMatrixTranspose.create(scope, input, type, options);
-  }
-
-  /**
-   * Creates an all-zeros CSRSparseMatrix with shape `dense_shape`.
-   *
-   * @param denseShape The desired matrix shape.
-   * @param type
-   * @return a new instance of SparseMatrixZeros
-   */
-  public <T extends TType> SparseMatrixZeros sparseMatrixZeros(Operand<TInt64> denseShape,
-      DataType<T> type) {
-    return SparseMatrixZeros.create(scope, denseShape, type);
-  }
-
-  /**
-   * Converts a SparseTensor to a (possibly batched) CSRSparseMatrix.
-   *
-   * @param indices SparseTensor indices.
-   * @param values SparseTensor values.
-   * @param denseShape SparseTensor dense shape.
-   * @return a new instance of SparseTensorToCSRSparseMatrix
-   */
-  public <T extends TType> SparseTensorToCSRSparseMatrix sparseTensorToCSRSparseMatrix(
-      Operand<TInt64> indices, Operand<T> values, Operand<TInt64> denseShape) {
-    return SparseTensorToCSRSparseMatrix.create(scope, indices, values, denseShape);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/MathOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/MathOps.java
index 1f08502ca44..d89679c5084 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/MathOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/MathOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.math.Abs;
@@ -289,7 +288,7 @@ public <T extends TType> Angle<TFloat32> angle(Operand<T> input) {
    * @param Tout
    * @return a new instance of Angle
    */
-  public <U extends TNumber, T extends TType> Angle<U> angle(Operand<T> input, DataType<U> Tout) {
+  public <U extends TNumber, T extends TType> Angle<U> angle(Operand<T> input, Class<U> Tout) {
     return Angle.create(scope, input, Tout);
   }
 
@@ -357,7 +356,7 @@ public <T extends TType, U extends TNumber> ArgMax<TInt64> argMax(Operand<T> inp
    * @return a new instance of ArgMax
    */
   public <V extends TNumber, T extends TType, U extends TNumber> ArgMax<V> argMax(Operand<T> input,
-      Operand<U> dimension, DataType<V> outputType) {
+      Operand<U> dimension, Class<V> outputType) {
     return ArgMax.create(scope, input, dimension, outputType);
   }
 
@@ -412,7 +411,7 @@ public <T extends TType, U extends TNumber> ArgMin<TInt64> argMin(Operand<T> inp
    * @return a new instance of ArgMin
    */
   public <V extends TNumber, T extends TType, U extends TNumber> ArgMin<V> argMin(Operand<T> input,
-      Operand<U> dimension, DataType<V> outputType) {
+      Operand<U> dimension, Class<V> outputType) {
     return ArgMin.create(scope, input, dimension, outputType);
   }
 
@@ -651,7 +650,7 @@ public <T extends TType> ComplexAbs<TFloat32> complexAbs(Operand<T> x) {
    * @return a new instance of ComplexAbs
    */
   public <U extends TNumber, T extends TType> ComplexAbs<U> complexAbs(Operand<T> x,
-      DataType<U> Tout) {
+      Class<U> Tout) {
     return ComplexAbs.create(scope, x, Tout);
   }
 
@@ -1178,7 +1177,7 @@ public <T extends TType> Imag<TFloat32> imag(Operand<T> input) {
    * @param Tout
    * @return a new instance of Imag
    */
-  public <U extends TNumber, T extends TType> Imag<U> imag(Operand<T> input, DataType<U> Tout) {
+  public <U extends TNumber, T extends TType> Imag<U> imag(Operand<T> input, Class<U> Tout) {
     return Imag.create(scope, input, Tout);
   }
 
@@ -1635,7 +1634,7 @@ public <T extends TType> Pow<T> pow(Operand<T> x, Operand<T> y) {
    */
   public <V extends TType, T extends TType, U extends TType> QuantizedAdd<V> quantizedAdd(
       Operand<T> x, Operand<U> y, Operand<TFloat32> minX, Operand<TFloat32> maxX,
-      Operand<TFloat32> minY, Operand<TFloat32> maxY, DataType<V> Toutput) {
+      Operand<TFloat32> minY, Operand<TFloat32> maxY, Class<V> Toutput) {
     return QuantizedAdd.create(scope, x, y, minX, maxX, minY, maxY, Toutput);
   }
 
@@ -1654,7 +1653,7 @@ public <V extends TType, T extends TType, U extends TType> QuantizedAdd<V> quant
    */
   public <V extends TType, T extends TType, U extends TType> QuantizedMul<V> quantizedMul(
       Operand<T> x, Operand<U> y, Operand<TFloat32> minX, Operand<TFloat32> maxX,
-      Operand<TFloat32> minY, Operand<TFloat32> maxY, DataType<V> Toutput) {
+      Operand<TFloat32> minY, Operand<TFloat32> maxY, Class<V> Toutput) {
     return QuantizedMul.create(scope, x, y, minX, maxX, minY, maxY, Toutput);
   }
 
@@ -1699,7 +1698,7 @@ public <T extends TType> Real<TFloat32> real(Operand<T> input) {
    * @param Tout
    * @return a new instance of Real
    */
-  public <U extends TNumber, T extends TType> Real<U> real(Operand<T> input, DataType<U> Tout) {
+  public <U extends TNumber, T extends TType> Real<U> real(Operand<T> input, Class<U> Tout) {
     return Real.create(scope, input, Tout);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/NnOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/NnOps.java
index 81a24514a08..0f523cf9dfb 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/NnOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/NnOps.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.nn.AvgPool;
 import org.tensorflow.op.nn.AvgPool3d;
@@ -642,8 +641,8 @@ public <T extends TNumber> CudnnRNNParamsToCanonical<T> cudnnRNNParamsToCanonica
    * @return a new instance of CudnnRnnParamsSize
    */
   public <U extends TNumber, T extends TNumber> CudnnRnnParamsSize<U> cudnnRnnParamsSize(
-      Operand<TInt32> numLayers, Operand<TInt32> numUnits, Operand<TInt32> inputSize, DataType<T> T,
-      DataType<U> S, CudnnRnnParamsSize.Options... options) {
+      Operand<TInt32> numLayers, Operand<TInt32> numUnits, Operand<TInt32> inputSize, Class<T> T,
+      Class<U> S, CudnnRnnParamsSize.Options... options) {
     return CudnnRnnParamsSize.create(scope, numLayers, numUnits, inputSize, T, S, options);
   }
 
@@ -1501,7 +1500,7 @@ public <T extends TNumber> MaxPoolWithArgmax<T, TInt64> maxPoolWithArgmax(Operan
    * @return a new instance of MaxPoolWithArgmax
    */
   public <T extends TNumber, U extends TNumber> MaxPoolWithArgmax<T, U> maxPoolWithArgmax(
-      Operand<T> input, List<Long> ksize, List<Long> strides, DataType<U> Targmax, String padding,
+      Operand<T> input, List<Long> ksize, List<Long> strides, Class<U> Targmax, String padding,
       MaxPoolWithArgmax.Options... options) {
     return MaxPoolWithArgmax.create(scope, input, ksize, strides, Targmax, padding, options);
   }
@@ -1588,7 +1587,7 @@ public <U extends TType, T extends TType> QuantizedBatchNormWithGlobalNormalizat
       Operand<T> t, Operand<TFloat32> tMin, Operand<TFloat32> tMax, Operand<T> m,
       Operand<TFloat32> mMin, Operand<TFloat32> mMax, Operand<T> v, Operand<TFloat32> vMin,
       Operand<TFloat32> vMax, Operand<T> beta, Operand<TFloat32> betaMin, Operand<TFloat32> betaMax,
-      Operand<T> gamma, Operand<TFloat32> gammaMin, Operand<TFloat32> gammaMax, DataType<U> outType,
+      Operand<T> gamma, Operand<TFloat32> gammaMin, Operand<TFloat32> gammaMax, Class<U> outType,
       Float varianceEpsilon, Boolean scaleAfterNormalization) {
     return QuantizedBatchNormWithGlobalNormalization.create(scope, t, tMin, tMax, m, mMin, mMax, v, vMin, vMax, beta, betaMin, betaMax, gamma, gammaMin, gammaMax, outType, varianceEpsilon, scaleAfterNormalization);
   }
@@ -1610,7 +1609,7 @@ public <U extends TType, T extends TType> QuantizedBatchNormWithGlobalNormalizat
    */
   public <V extends TType, T extends TType, U extends TType> QuantizedBiasAdd<V> quantizedBiasAdd(
       Operand<T> input, Operand<U> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput,
-      Operand<TFloat32> minBias, Operand<TFloat32> maxBias, DataType<V> outType) {
+      Operand<TFloat32> minBias, Operand<TFloat32> maxBias, Class<V> outType) {
     return QuantizedBiasAdd.create(scope, input, bias, minInput, maxInput, minBias, maxBias, outType);
   }
 
@@ -1638,7 +1637,7 @@ public <V extends TType, T extends TType, U extends TType> QuantizedBiasAdd<V> q
    */
   public <V extends TType, T extends TType, U extends TType> QuantizedConv2d<V> quantizedConv2d(
       Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput,
-      Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType,
+      Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType,
       List<Long> strides, String padding, QuantizedConv2d.Options... options) {
     return QuantizedConv2d.create(scope, input, filter, minInput, maxInput, minFilter, maxFilter, outType, strides, padding, options);
   }
@@ -1689,7 +1688,7 @@ public <T extends TType> QuantizedMaxPool<T> quantizedMaxPool(Operand<T> input,
    * @return a new instance of QuantizedRelu
    */
   public <U extends TType, T extends TType> QuantizedRelu<U> quantizedRelu(Operand<T> features,
-      Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, DataType<U> outType) {
+      Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, Class<U> outType) {
     return QuantizedRelu.create(scope, features, minFeatures, maxFeatures, outType);
   }
 
@@ -1704,7 +1703,7 @@ public <U extends TType, T extends TType> QuantizedRelu<U> quantizedRelu(Operand
    * @return a new instance of QuantizedRelu6
    */
   public <U extends TType, T extends TType> QuantizedRelu6<U> quantizedRelu6(Operand<T> features,
-      Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, DataType<U> outType) {
+      Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, Class<U> outType) {
     return QuantizedRelu6.create(scope, features, minFeatures, maxFeatures, outType);
   }
 
@@ -1721,7 +1720,7 @@ public <U extends TType, T extends TType> QuantizedRelu6<U> quantizedRelu6(Opera
    */
   public <U extends TType, T extends TType> QuantizedReluX<U> quantizedReluX(Operand<T> features,
       Operand<TFloat32> maxValue, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures,
-      DataType<U> outType) {
+      Class<U> outType) {
     return QuantizedReluX.create(scope, features, maxValue, minFeatures, maxFeatures, outType);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/Ops.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/Ops.java
index 8c7aa1c0408..f1abd40eb39 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/Ops.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/Ops.java
@@ -19,11 +19,9 @@
 
 import java.nio.charset.Charset;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.EagerSession;
 import org.tensorflow.ExecutionEnvironment;
 import org.tensorflow.Operand;
-import org.tensorflow.Tensor;
 import org.tensorflow.ndarray.BooleanNdArray;
 import org.tensorflow.ndarray.ByteNdArray;
 import org.tensorflow.ndarray.DoubleNdArray;
@@ -243,8 +241,6 @@
 import org.tensorflow.op.core.TensorListSetItem;
 import org.tensorflow.op.core.TensorListSplit;
 import org.tensorflow.op.core.TensorListStack;
-import org.tensorflow.op.core.TensorScatterMax;
-import org.tensorflow.op.core.TensorScatterMin;
 import org.tensorflow.op.core.TensorScatterNdAdd;
 import org.tensorflow.op.core.TensorScatterNdMax;
 import org.tensorflow.op.core.TensorScatterNdMin;
@@ -647,7 +643,7 @@ public <T extends TType> AssignVariableOp assignVariableOp(Operand<?> resource,
    * @param options carries optional attributes values
    * @return a new instance of Barrier
    */
-  public Barrier barrier(List<DataType<?>> componentTypes, Barrier.Options... options) {
+  public Barrier barrier(List<Class<? extends TType>> componentTypes, Barrier.Options... options) {
     return Barrier.create(scope, componentTypes, options);
   }
 
@@ -728,7 +724,7 @@ public BarrierReadySize barrierReadySize(Operand<TString> handle) {
    * @return a new instance of BarrierTakeMany
    */
   public BarrierTakeMany barrierTakeMany(Operand<TString> handle, Operand<TInt32> numElements,
-      List<DataType<?>> componentTypes, BarrierTakeMany.Options... options) {
+      List<Class<? extends TType>> componentTypes, BarrierTakeMany.Options... options) {
     return BarrierTakeMany.create(scope, handle, numElements, componentTypes, options);
   }
 
@@ -988,7 +984,7 @@ public <T extends TType, U extends TNumber, V extends TNumber> BatchToSpaceNd<T>
    * @param type
    * @return a new instance of Bitcast
    */
-  public <U extends TType, T extends TType> Bitcast<U> bitcast(Operand<T> input, DataType<U> type) {
+  public <U extends TType, T extends TType> Bitcast<U> bitcast(Operand<T> input, Class<U> type) {
     return Bitcast.create(scope, input, type);
   }
 
@@ -1713,7 +1709,7 @@ public Constant<TInt64> constant(Shape shape) {
    * @param tensor a Tensor holding the constant value
    * @return a constant of the same data type as `tensor`
    */
-  public <T extends TType> Constant<T> constant(Tensor<T> tensor) {
+  public <T extends TType> Constant<T> constant(T tensor) {
     return Constant.create(scope, tensor);
   }
 
@@ -1865,15 +1861,14 @@ public Constant<TString> constant(Charset charset, Shape shape, DataBuffer<Strin
    * Create a constant with data from the given buffer.
    *
    * @param scope is a scope used to add the underlying operation.
-   * @param type the tensor datatype.
+   * @param type the tensor type.
    * @param shape the tensor shape.
    * @param data a buffer containing the tensor data.
    * @return a constant of type `type`
    * @throws IllegalArgumentException If the tensor datatype or shape is not compatible with the
    *      buffer
    */
-  public <T extends TType> Constant<T> constant(DataType<T> type, Shape shape,
-      ByteDataBuffer data) {
+  public <T extends TType> Constant<T> constant(Class<T> type, Shape shape, ByteDataBuffer data) {
     return Constant.tensorOf(scope, type, shape, data);
   }
 
@@ -2134,7 +2129,7 @@ public <T extends TType> EditDistance editDistance(Operand<TInt64> hypothesisInd
    * @param options carries optional attributes values
    * @return a new instance of Empty
    */
-  public <T extends TType> Empty<T> empty(Operand<TInt32> shape, DataType<T> dtype,
+  public <T extends TType> Empty<T> empty(Operand<TInt32> shape, Class<T> dtype,
       Empty.Options... options) {
     return Empty.create(scope, shape, dtype, options);
   }
@@ -2155,7 +2150,7 @@ public <T extends TType> Empty<T> empty(Operand<TInt32> shape, DataType<T> dtype
    * @return a new instance of EmptyTensorList
    */
   public <T extends TNumber, U extends TType> EmptyTensorList emptyTensorList(
-      Operand<T> elementShape, Operand<TInt32> maxNumElements, DataType<U> elementDtype) {
+      Operand<T> elementShape, Operand<TInt32> maxNumElements, Class<U> elementDtype) {
     return EmptyTensorList.create(scope, elementShape, maxNumElements, elementDtype);
   }
 
@@ -2487,7 +2482,7 @@ public <T extends TType> GetSessionHandle getSessionHandle(Operand<T> value) {
    * @return a new instance of GetSessionTensor
    */
   public <T extends TType> GetSessionTensor<T> getSessionTensor(Operand<TString> handle,
-      DataType<T> dtype) {
+      Class<T> dtype) {
     return GetSessionTensor.create(scope, handle, dtype);
   }
 
@@ -2567,8 +2562,8 @@ public <T extends TType> GuaranteeConst<T> guaranteeConst(Operand<T> input) {
    * @param options carries optional attributes values
    * @return a new instance of HashTable
    */
-  public <T extends TType, U extends TType> HashTable hashTable(DataType<T> keyDtype,
-      DataType<U> valueDtype, HashTable.Options... options) {
+  public <T extends TType, U extends TType> HashTable hashTable(Class<T> keyDtype,
+      Class<U> valueDtype, HashTable.Options... options) {
     return HashTable.create(scope, keyDtype, valueDtype, options);
   }
 
@@ -2631,7 +2626,7 @@ public <T extends TNumber> HistogramFixedWidth<TInt32> histogramFixedWidth(Opera
    * @return a new instance of HistogramFixedWidth
    */
   public <U extends TNumber, T extends TNumber> HistogramFixedWidth<U> histogramFixedWidth(
-      Operand<T> values, Operand<T> valueRange, Operand<TInt32> nbins, DataType<U> dtype) {
+      Operand<T> values, Operand<T> valueRange, Operand<TInt32> nbins, Class<U> dtype) {
     return HistogramFixedWidth.create(scope, values, valueRange, nbins, dtype);
   }
 
@@ -2681,7 +2676,7 @@ public IdentityN identityN(Iterable<Operand<?>> input) {
    *  NewReadOnlyMemoryRegionFromFile in tensorflow::Env.
    * @return a new instance of ImmutableConst
    */
-  public <T extends TType> ImmutableConst<T> immutableConst(DataType<T> dtype, Shape shape,
+  public <T extends TType> ImmutableConst<T> immutableConst(Class<T> dtype, Shape shape,
       String memoryRegionName) {
     return ImmutableConst.create(scope, dtype, shape, memoryRegionName);
   }
@@ -2876,7 +2871,7 @@ public <T extends TType> IsVariableInitialized isVariableInitialized(Operand<T>
    * @return a new instance of LookupTableExport
    */
   public <T extends TType, U extends TType> LookupTableExport<T, U> lookupTableExport(
-      Operand<?> tableHandle, DataType<T> Tkeys, DataType<U> Tvalues) {
+      Operand<?> tableHandle, Class<T> Tkeys, Class<U> Tvalues) {
     return LookupTableExport.create(scope, tableHandle, Tkeys, Tvalues);
   }
 
@@ -2962,7 +2957,7 @@ public LoopCond loopCond(Operand<TBool> input) {
    * @param options carries optional attributes values
    * @return a new instance of MapClear
    */
-  public MapClear mapClear(List<DataType<?>> dtypes, MapClear.Options... options) {
+  public MapClear mapClear(List<Class<? extends TType>> dtypes, MapClear.Options... options) {
     return MapClear.create(scope, dtypes, options);
   }
 
@@ -2973,7 +2968,7 @@ public MapClear mapClear(List<DataType<?>> dtypes, MapClear.Options... options)
    * @param options carries optional attributes values
    * @return a new instance of MapIncompleteSize
    */
-  public MapIncompleteSize mapIncompleteSize(List<DataType<?>> dtypes,
+  public MapIncompleteSize mapIncompleteSize(List<Class<? extends TType>> dtypes,
       MapIncompleteSize.Options... options) {
     return MapIncompleteSize.create(scope, dtypes, options);
   }
@@ -2990,8 +2985,8 @@ public MapIncompleteSize mapIncompleteSize(List<DataType<?>> dtypes,
    * @param options carries optional attributes values
    * @return a new instance of MapPeek
    */
-  public MapPeek mapPeek(Operand<TInt64> key, Operand<TInt32> indices, List<DataType<?>> dtypes,
-      MapPeek.Options... options) {
+  public MapPeek mapPeek(Operand<TInt64> key, Operand<TInt32> indices,
+      List<Class<? extends TType>> dtypes, MapPeek.Options... options) {
     return MapPeek.create(scope, key, indices, dtypes, options);
   }
 
@@ -3002,7 +2997,7 @@ public MapPeek mapPeek(Operand<TInt64> key, Operand<TInt32> indices, List<DataTy
    * @param options carries optional attributes values
    * @return a new instance of MapSize
    */
-  public MapSize mapSize(List<DataType<?>> dtypes, MapSize.Options... options) {
+  public MapSize mapSize(List<Class<? extends TType>> dtypes, MapSize.Options... options) {
     return MapSize.create(scope, dtypes, options);
   }
 
@@ -3018,7 +3013,8 @@ public MapSize mapSize(List<DataType<?>> dtypes, MapSize.Options... options) {
    * @return a new instance of MapStage
    */
   public MapStage mapStage(Operand<TInt64> key, Operand<TInt32> indices,
-      Iterable<Operand<?>> values, List<DataType<?>> dtypes, MapStage.Options... options) {
+      Iterable<Operand<?>> values, List<Class<? extends TType>> dtypes,
+      MapStage.Options... options) {
     return MapStage.create(scope, key, indices, values, dtypes, options);
   }
 
@@ -3035,7 +3031,7 @@ public MapStage mapStage(Operand<TInt64> key, Operand<TInt32> indices,
    * @return a new instance of MapUnstage
    */
   public MapUnstage mapUnstage(Operand<TInt64> key, Operand<TInt32> indices,
-      List<DataType<?>> dtypes, MapUnstage.Options... options) {
+      List<Class<? extends TType>> dtypes, MapUnstage.Options... options) {
     return MapUnstage.create(scope, key, indices, dtypes, options);
   }
 
@@ -3050,8 +3046,8 @@ public MapUnstage mapUnstage(Operand<TInt64> key, Operand<TInt32> indices,
    * @param options carries optional attributes values
    * @return a new instance of MapUnstageNoKey
    */
-  public MapUnstageNoKey mapUnstageNoKey(Operand<TInt32> indices, List<DataType<?>> dtypes,
-      MapUnstageNoKey.Options... options) {
+  public MapUnstageNoKey mapUnstageNoKey(Operand<TInt32> indices,
+      List<Class<? extends TType>> dtypes, MapUnstageNoKey.Options... options) {
     return MapUnstageNoKey.create(scope, indices, dtypes, options);
   }
 
@@ -3192,7 +3188,7 @@ public <T extends TType, U extends TNumber> MirrorPad<T> mirrorPad(Operand<T> in
    * @return a new instance of MlirPassthroughOp
    */
   public MlirPassthroughOp mlirPassthroughOp(Iterable<Operand<?>> inputs, String mlirModule,
-      List<DataType<?>> Toutputs) {
+      List<Class<? extends TType>> Toutputs) {
     return MlirPassthroughOp.create(scope, inputs, mlirModule, Toutputs);
   }
 
@@ -3214,7 +3210,7 @@ public MlirPassthroughOp mlirPassthroughOp(Iterable<Operand<?>> inputs, String m
    * @return a new instance of MutableDenseHashTable
    */
   public <T extends TType, U extends TType> MutableDenseHashTable mutableDenseHashTable(
-      Operand<T> emptyKey, Operand<T> deletedKey, DataType<U> valueDtype,
+      Operand<T> emptyKey, Operand<T> deletedKey, Class<U> valueDtype,
       MutableDenseHashTable.Options... options) {
     return MutableDenseHashTable.create(scope, emptyKey, deletedKey, valueDtype, options);
   }
@@ -3231,8 +3227,8 @@ public <T extends TType, U extends TType> MutableDenseHashTable mutableDenseHash
    * @param options carries optional attributes values
    * @return a new instance of MutableHashTable
    */
-  public <T extends TType, U extends TType> MutableHashTable mutableHashTable(DataType<T> keyDtype,
-      DataType<U> valueDtype, MutableHashTable.Options... options) {
+  public <T extends TType, U extends TType> MutableHashTable mutableHashTable(Class<T> keyDtype,
+      Class<U> valueDtype, MutableHashTable.Options... options) {
     return MutableHashTable.create(scope, keyDtype, valueDtype, options);
   }
 
@@ -3249,7 +3245,7 @@ public <T extends TType, U extends TType> MutableHashTable mutableHashTable(Data
    * @return a new instance of MutableHashTableOfTensors
    */
   public <T extends TType, U extends TType> MutableHashTableOfTensors mutableHashTableOfTensors(
-      DataType<T> keyDtype, DataType<U> valueDtype, MutableHashTableOfTensors.Options... options) {
+      Class<T> keyDtype, Class<U> valueDtype, MutableHashTableOfTensors.Options... options) {
     return MutableHashTableOfTensors.create(scope, keyDtype, valueDtype, options);
   }
 
@@ -3443,7 +3439,7 @@ public <T extends TType> OnesLike<T> onesLike(Operand<T> x) {
    * @param options carries optional attributes values
    * @return a new instance of OrderedMapClear
    */
-  public OrderedMapClear orderedMapClear(List<DataType<?>> dtypes,
+  public OrderedMapClear orderedMapClear(List<Class<? extends TType>> dtypes,
       OrderedMapClear.Options... options) {
     return OrderedMapClear.create(scope, dtypes, options);
   }
@@ -3455,7 +3451,7 @@ public OrderedMapClear orderedMapClear(List<DataType<?>> dtypes,
    * @param options carries optional attributes values
    * @return a new instance of OrderedMapIncompleteSize
    */
-  public OrderedMapIncompleteSize orderedMapIncompleteSize(List<DataType<?>> dtypes,
+  public OrderedMapIncompleteSize orderedMapIncompleteSize(List<Class<? extends TType>> dtypes,
       OrderedMapIncompleteSize.Options... options) {
     return OrderedMapIncompleteSize.create(scope, dtypes, options);
   }
@@ -3474,7 +3470,7 @@ public OrderedMapIncompleteSize orderedMapIncompleteSize(List<DataType<?>> dtype
    * @return a new instance of OrderedMapPeek
    */
   public OrderedMapPeek orderedMapPeek(Operand<TInt64> key, Operand<TInt32> indices,
-      List<DataType<?>> dtypes, OrderedMapPeek.Options... options) {
+      List<Class<? extends TType>> dtypes, OrderedMapPeek.Options... options) {
     return OrderedMapPeek.create(scope, key, indices, dtypes, options);
   }
 
@@ -3485,7 +3481,7 @@ public OrderedMapPeek orderedMapPeek(Operand<TInt64> key, Operand<TInt32> indice
    * @param options carries optional attributes values
    * @return a new instance of OrderedMapSize
    */
-  public OrderedMapSize orderedMapSize(List<DataType<?>> dtypes,
+  public OrderedMapSize orderedMapSize(List<Class<? extends TType>> dtypes,
       OrderedMapSize.Options... options) {
     return OrderedMapSize.create(scope, dtypes, options);
   }
@@ -3504,7 +3500,8 @@ public OrderedMapSize orderedMapSize(List<DataType<?>> dtypes,
    * @return a new instance of OrderedMapStage
    */
   public OrderedMapStage orderedMapStage(Operand<TInt64> key, Operand<TInt32> indices,
-      Iterable<Operand<?>> values, List<DataType<?>> dtypes, OrderedMapStage.Options... options) {
+      Iterable<Operand<?>> values, List<Class<? extends TType>> dtypes,
+      OrderedMapStage.Options... options) {
     return OrderedMapStage.create(scope, key, indices, values, dtypes, options);
   }
 
@@ -3521,7 +3518,7 @@ public OrderedMapStage orderedMapStage(Operand<TInt64> key, Operand<TInt32> indi
    * @return a new instance of OrderedMapUnstage
    */
   public OrderedMapUnstage orderedMapUnstage(Operand<TInt64> key, Operand<TInt32> indices,
-      List<DataType<?>> dtypes, OrderedMapUnstage.Options... options) {
+      List<Class<? extends TType>> dtypes, OrderedMapUnstage.Options... options) {
     return OrderedMapUnstage.create(scope, key, indices, dtypes, options);
   }
 
@@ -3537,7 +3534,7 @@ public OrderedMapUnstage orderedMapUnstage(Operand<TInt64> key, Operand<TInt32>
    * @return a new instance of OrderedMapUnstageNoKey
    */
   public OrderedMapUnstageNoKey orderedMapUnstageNoKey(Operand<TInt32> indices,
-      List<DataType<?>> dtypes, OrderedMapUnstageNoKey.Options... options) {
+      List<Class<? extends TType>> dtypes, OrderedMapUnstageNoKey.Options... options) {
     return OrderedMapUnstageNoKey.create(scope, indices, dtypes, options);
   }
 
@@ -3688,7 +3685,7 @@ public <T extends TType> ParallelDynamicStitch<T> parallelDynamicStitch(
    * @param options carries optional attributes values
    * @return a new instance of Placeholder
    */
-  public <T extends TType> Placeholder<T> placeholder(DataType<T> dtype,
+  public <T extends TType> Placeholder<T> placeholder(Class<T> dtype,
       Placeholder.Options... options) {
     return Placeholder.create(scope, dtype, options);
   }
@@ -3817,8 +3814,7 @@ public <T extends TType> Rank rank(Operand<T> input) {
    * @param dtype the dtype of the value.
    * @return a new instance of ReadVariableOp
    */
-  public <T extends TType> ReadVariableOp<T> readVariableOp(Operand<?> resource,
-      DataType<T> dtype) {
+  public <T extends TType> ReadVariableOp<T> readVariableOp(Operand<?> resource, Class<T> dtype) {
     return ReadVariableOp.create(scope, resource, dtype);
   }
 
@@ -3999,7 +3995,7 @@ public <T extends TType> RefSwitch<T> refSwitch(Operand<T> data, Operand<TBool>
    * @return a new instance of RemoteFusedGraphExecute
    */
   public RemoteFusedGraphExecute remoteFusedGraphExecute(Iterable<Operand<?>> inputs,
-      List<DataType<?>> Toutputs, String serializedRemoteFusedGraphExecuteInfo) {
+      List<Class<? extends TType>> Toutputs, String serializedRemoteFusedGraphExecuteInfo) {
     return RemoteFusedGraphExecute.create(scope, inputs, Toutputs, serializedRemoteFusedGraphExecuteInfo);
   }
 
@@ -4086,7 +4082,7 @@ public <T extends TType, U extends TNumber> Reshape<T> reshape(Operand<T> tensor
    * @return a new instance of ResourceCountUpTo
    */
   public <T extends TNumber> ResourceCountUpTo<T> resourceCountUpTo(Operand<?> resource, Long limit,
-      DataType<T> T) {
+      Class<T> T) {
     return ResourceCountUpTo.create(scope, resource, limit, T);
   }
 
@@ -4114,7 +4110,7 @@ public <T extends TNumber> ResourceCountUpTo<T> resourceCountUpTo(Operand<?> res
    * @return a new instance of ResourceGather
    */
   public <U extends TType, T extends TNumber> ResourceGather<U> resourceGather(Operand<?> resource,
-      Operand<T> indices, DataType<U> dtype, ResourceGather.Options... options) {
+      Operand<T> indices, Class<U> dtype, ResourceGather.Options... options) {
     return ResourceGather.create(scope, resource, indices, dtype, options);
   }
 
@@ -4127,7 +4123,7 @@ public <U extends TType, T extends TNumber> ResourceGather<U> resourceGather(Ope
    * @return a new instance of ResourceGatherNd
    */
   public <U extends TType, T extends TNumber> ResourceGatherNd<U> resourceGatherNd(
-      Operand<?> resource, Operand<T> indices, DataType<U> dtype) {
+      Operand<?> resource, Operand<T> indices, Class<U> dtype) {
     return ResourceGatherNd.create(scope, resource, indices, dtype);
   }
 
@@ -5406,7 +5402,7 @@ public <T extends TType> SetDiff1d<T, TInt32> setDiff1d(Operand<T> x, Operand<T>
    * @return a new instance of SetDiff1d
    */
   public <T extends TType, U extends TNumber> SetDiff1d<T, U> setDiff1d(Operand<T> x, Operand<T> y,
-      DataType<U> outIdx) {
+      Class<U> outIdx) {
     return SetDiff1d.create(scope, x, y, outIdx);
   }
 
@@ -5467,7 +5463,7 @@ public <T extends TType> org.tensorflow.op.core.Shape<TInt32> shape(Operand<T> i
    * @return a new instance of Shape
    */
   public <U extends TNumber, T extends TType> org.tensorflow.op.core.Shape<U> shape(
-      Operand<T> input, DataType<U> outType) {
+      Operand<T> input, Class<U> outType) {
     return org.tensorflow.op.core.Shape.create(scope, input, outType);
   }
 
@@ -5495,7 +5491,7 @@ public <T extends TType> ShapeN<TInt32> shapeN(Iterable<Operand<T>> input) {
    * @return a new instance of ShapeN
    */
   public <U extends TNumber, T extends TType> ShapeN<U> shapeN(Iterable<Operand<T>> input,
-      DataType<U> outType) {
+      Class<U> outType) {
     return ShapeN.create(scope, input, outType);
   }
 
@@ -5536,7 +5532,7 @@ public <T extends TType> Size<TInt32> size(Operand<T> input) {
    * @param outType
    * @return a new instance of Size
    */
-  public <U extends TNumber, T extends TType> Size<U> size(Operand<T> input, DataType<U> outType) {
+  public <U extends TNumber, T extends TType> Size<U> size(Operand<T> input, Class<U> outType) {
     return Size.create(scope, input, outType);
   }
 
@@ -5816,7 +5812,7 @@ public Stage stage(Iterable<Operand<?>> values, Stage.Options... options) {
    * @param options carries optional attributes values
    * @return a new instance of StageClear
    */
-  public StageClear stageClear(List<DataType<?>> dtypes, StageClear.Options... options) {
+  public StageClear stageClear(List<Class<? extends TType>> dtypes, StageClear.Options... options) {
     return StageClear.create(scope, dtypes, options);
   }
 
@@ -5832,7 +5828,7 @@ public StageClear stageClear(List<DataType<?>> dtypes, StageClear.Options... opt
    * @param options carries optional attributes values
    * @return a new instance of StagePeek
    */
-  public StagePeek stagePeek(Operand<TInt32> index, List<DataType<?>> dtypes,
+  public StagePeek stagePeek(Operand<TInt32> index, List<Class<? extends TType>> dtypes,
       StagePeek.Options... options) {
     return StagePeek.create(scope, index, dtypes, options);
   }
@@ -5844,7 +5840,7 @@ public StagePeek stagePeek(Operand<TInt32> index, List<DataType<?>> dtypes,
    * @param options carries optional attributes values
    * @return a new instance of StageSize
    */
-  public StageSize stageSize(List<DataType<?>> dtypes, StageSize.Options... options) {
+  public StageSize stageSize(List<Class<? extends TType>> dtypes, StageSize.Options... options) {
     return StageSize.create(scope, dtypes, options);
   }
 
@@ -6109,7 +6105,7 @@ public <T extends TType> SwitchCond<T> switchCond(Operand<T> data, Operand<TBool
    * @param options carries optional attributes values
    * @return a new instance of TemporaryVariable
    */
-  public <T extends TType> TemporaryVariable<T> temporaryVariable(Shape shape, DataType<T> dtype,
+  public <T extends TType> TemporaryVariable<T> temporaryVariable(Shape shape, Class<T> dtype,
       TemporaryVariable.Options... options) {
     return TemporaryVariable.create(scope, shape, dtype, options);
   }
@@ -6124,7 +6120,7 @@ public <T extends TType> TemporaryVariable<T> temporaryVariable(Shape shape, Dat
    * @param options carries optional attributes values
    * @return a new instance of TensorArray
    */
-  public <T extends TType> TensorArray tensorArray(Operand<TInt32> size, DataType<T> dtype,
+  public <T extends TType> TensorArray tensorArray(Operand<TInt32> size, Class<T> dtype,
       TensorArray.Options... options) {
     return TensorArray.create(scope, size, dtype, options);
   }
@@ -6164,7 +6160,7 @@ public TensorArrayClose tensorArrayClose(Operand<?> handle) {
    * @return a new instance of TensorArrayConcat
    */
   public <T extends TType> TensorArrayConcat<T> tensorArrayConcat(Operand<?> handle,
-      Operand<TFloat32> flowIn, DataType<T> dtype, TensorArrayConcat.Options... options) {
+      Operand<TFloat32> flowIn, Class<T> dtype, TensorArrayConcat.Options... options) {
     return TensorArrayConcat.create(scope, handle, flowIn, dtype, options);
   }
 
@@ -6182,7 +6178,7 @@ public <T extends TType> TensorArrayConcat<T> tensorArrayConcat(Operand<?> handl
    * @return a new instance of TensorArrayGather
    */
   public <T extends TType> TensorArrayGather<T> tensorArrayGather(Operand<?> handle,
-      Operand<TInt32> indices, Operand<TFloat32> flowIn, DataType<T> dtype,
+      Operand<TInt32> indices, Operand<TFloat32> flowIn, Class<T> dtype,
       TensorArrayGather.Options... options) {
     return TensorArrayGather.create(scope, handle, indices, flowIn, dtype, options);
   }
@@ -6270,7 +6266,7 @@ public TensorArrayGradWithShape tensorArrayGradWithShape(Operand<?> handle,
    * @return a new instance of TensorArrayPack
    */
   public <T extends TType> TensorArrayPack<T> tensorArrayPack(Operand<TString> handle,
-      Operand<TFloat32> flowIn, DataType<T> dtype, TensorArrayPack.Options... options) {
+      Operand<TFloat32> flowIn, Class<T> dtype, TensorArrayPack.Options... options) {
     return TensorArrayPack.create(scope, handle, flowIn, dtype, options);
   }
 
@@ -6285,7 +6281,7 @@ public <T extends TType> TensorArrayPack<T> tensorArrayPack(Operand<TString> han
    * @return a new instance of TensorArrayRead
    */
   public <T extends TType> TensorArrayRead<T> tensorArrayRead(Operand<?> handle,
-      Operand<TInt32> index, Operand<TFloat32> flowIn, DataType<T> dtype) {
+      Operand<TInt32> index, Operand<TFloat32> flowIn, Class<T> dtype) {
     return TensorArrayRead.create(scope, handle, index, flowIn, dtype);
   }
 
@@ -6402,7 +6398,7 @@ public <T extends TType> TensorArrayWrite tensorArrayWrite(Operand<?> handle,
    */
   public <U extends TType, T extends TNumber> TensorListConcat<U> tensorListConcat(
       Operand<?> inputHandle, Operand<T> elementShape, Operand<TInt64> leadingDims,
-      DataType<U> elementDtype) {
+      Class<U> elementDtype) {
     return TensorListConcat.create(scope, inputHandle, elementShape, leadingDims, elementDtype);
   }
 
@@ -6414,7 +6410,7 @@ public <U extends TType, T extends TNumber> TensorListConcat<U> tensorListConcat
    * @return a new instance of TensorListConcatLists
    */
   public <T extends TType> TensorListConcatLists tensorListConcatLists(Operand<?> inputA,
-      Operand<?> inputB, DataType<T> elementDtype) {
+      Operand<?> inputB, Class<T> elementDtype) {
     return TensorListConcatLists.create(scope, inputA, inputB, elementDtype);
   }
 
@@ -6430,7 +6426,7 @@ public <T extends TType> TensorListConcatLists tensorListConcatLists(Operand<?>
    * @return a new instance of TensorListElementShape
    */
   public <T extends TNumber> TensorListElementShape<T> tensorListElementShape(
-      Operand<?> inputHandle, DataType<T> shapeType) {
+      Operand<?> inputHandle, Class<T> shapeType) {
     return TensorListElementShape.create(scope, inputHandle, shapeType);
   }
 
@@ -6469,7 +6465,7 @@ public <T extends TType, U extends TNumber> TensorListFromTensor tensorListFromT
    * @return a new instance of TensorListGather
    */
   public <T extends TType> TensorListGather<T> tensorListGather(Operand<?> inputHandle,
-      Operand<TInt32> indices, Operand<TInt32> elementShape, DataType<T> elementDtype) {
+      Operand<TInt32> indices, Operand<TInt32> elementShape, Class<T> elementDtype) {
     return TensorListGather.create(scope, inputHandle, indices, elementShape, elementDtype);
   }
 
@@ -6483,7 +6479,7 @@ public <T extends TType> TensorListGather<T> tensorListGather(Operand<?> inputHa
    * @return a new instance of TensorListGetItem
    */
   public <T extends TType> TensorListGetItem<T> tensorListGetItem(Operand<?> inputHandle,
-      Operand<TInt32> index, Operand<TInt32> elementShape, DataType<T> elementDtype) {
+      Operand<TInt32> index, Operand<TInt32> elementShape, Class<T> elementDtype) {
     return TensorListGetItem.create(scope, inputHandle, index, elementShape, elementDtype);
   }
 
@@ -6517,7 +6513,7 @@ public TensorListLength tensorListLength(Operand<?> inputHandle) {
    * @return a new instance of TensorListPopBack
    */
   public <T extends TType> TensorListPopBack<T> tensorListPopBack(Operand<?> inputHandle,
-      Operand<TInt32> elementShape, DataType<T> elementDtype) {
+      Operand<TInt32> elementShape, Class<T> elementDtype) {
     return TensorListPopBack.create(scope, inputHandle, elementShape, elementDtype);
   }
 
@@ -6564,7 +6560,7 @@ public <T extends TType> TensorListPushBackBatch tensorListPushBackBatch(Operand
    * @return a new instance of TensorListReserve
    */
   public <T extends TNumber, U extends TType> TensorListReserve tensorListReserve(
-      Operand<T> elementShape, Operand<TInt32> numElements, DataType<U> elementDtype) {
+      Operand<T> elementShape, Operand<TInt32> numElements, Class<U> elementDtype) {
     return TensorListReserve.create(scope, elementShape, numElements, elementDtype);
   }
 
@@ -6680,36 +6676,10 @@ public <T extends TType, U extends TNumber> TensorListSplit tensorListSplit(Oper
    * @return a new instance of TensorListStack
    */
   public <T extends TType> TensorListStack<T> tensorListStack(Operand<?> inputHandle,
-      Operand<TInt32> elementShape, DataType<T> elementDtype, TensorListStack.Options... options) {
+      Operand<TInt32> elementShape, Class<T> elementDtype, TensorListStack.Options... options) {
     return TensorListStack.create(scope, inputHandle, elementShape, elementDtype, options);
   }
 
-  /**
-   *
-   * @param <T> data type for {@code output()} output
-   * @param tensor Tensor to update.
-   * @param indices Index tensor.
-   * @param updates Updates to scatter into output.
-   * @return a new instance of TensorScatterMax
-   */
-  public <T extends TType, U extends TNumber> TensorScatterMax<T> tensorScatterMax(
-      Operand<T> tensor, Operand<U> indices, Operand<T> updates) {
-    return TensorScatterMax.create(scope, tensor, indices, updates);
-  }
-
-  /**
-   *
-   * @param <T> data type for {@code output()} output
-   * @param tensor Tensor to update.
-   * @param indices Index tensor.
-   * @param updates Updates to scatter into output.
-   * @return a new instance of TensorScatterMin
-   */
-  public <T extends TType, U extends TNumber> TensorScatterMin<T> tensorScatterMin(
-      Operand<T> tensor, Operand<U> indices, Operand<T> updates) {
-    return TensorScatterMin.create(scope, tensor, indices, updates);
-  }
-
   /**
    * Adds sparse `updates` to an existing tensor according to `indices`.
    *  <p>
@@ -7287,7 +7257,7 @@ public <T extends TType, U extends TNumber> Unique<T, TInt32> unique(Operand<T>
    * @return a new instance of Unique
    */
   public <T extends TType, V extends TNumber, U extends TNumber> Unique<T, V> unique(Operand<T> x,
-      Operand<U> axis, DataType<V> outIdx) {
+      Operand<U> axis, Class<V> outIdx) {
     return Unique.create(scope, x, axis, outIdx);
   }
 
@@ -7404,7 +7374,7 @@ public <T extends TType, U extends TNumber> UniqueWithCounts<T, TInt32> uniqueWi
    * @return a new instance of UniqueWithCounts
    */
   public <T extends TType, V extends TNumber, U extends TNumber> UniqueWithCounts<T, V> uniqueWithCounts(
-      Operand<T> x, Operand<U> axis, DataType<V> outIdx) {
+      Operand<T> x, Operand<U> axis, Class<V> outIdx) {
     return UniqueWithCounts.create(scope, x, axis, outIdx);
   }
 
@@ -7477,7 +7447,7 @@ public <T extends TType> Unstack<T> unstack(Operand<T> value, Long num,
    * @param options carries optional attributes values
    * @return a new instance of Unstage
    */
-  public Unstage unstage(List<DataType<?>> dtypes, Unstage.Options... options) {
+  public Unstage unstage(List<Class<? extends TType>> dtypes, Unstage.Options... options) {
     return Unstage.create(scope, dtypes, options);
   }
 
@@ -7490,7 +7460,7 @@ public Unstage unstage(List<DataType<?>> dtypes, Unstage.Options... options) {
    * @param options carries optional attributes values
    * @return a new instance of VarHandleOp
    */
-  public <T extends TType> VarHandleOp varHandleOp(DataType<T> dtype, Shape shape,
+  public <T extends TType> VarHandleOp varHandleOp(Class<T> dtype, Shape shape,
       VarHandleOp.Options... options) {
     return VarHandleOp.create(scope, dtype, shape, options);
   }
@@ -7533,7 +7503,7 @@ public <T extends TType> Variable<T> variable(Operand<T> init, Variable.Options.
    * @param options carries optional attributes values
    * @return a new instance of Variable
    */
-  public <T extends TType> Variable<T> variable(Shape shape, DataType<T> dtype,
+  public <T extends TType> Variable<T> variable(Shape shape, Class<T> dtype,
       Variable.Options... options) {
     return Variable.create(scope, shape, dtype, options);
   }
@@ -7573,7 +7543,7 @@ public VariableShape<TInt32> variableShape(Operand<?> input) {
    * @param outType
    * @return a new instance of VariableShape
    */
-  public <T extends TNumber> VariableShape<T> variableShape(Operand<?> input, DataType<T> outType) {
+  public <T extends TNumber> VariableShape<T> variableShape(Operand<?> input, Class<T> outType) {
     return VariableShape.create(scope, input, outType);
   }
 
@@ -7691,7 +7661,7 @@ public <T extends TType> XlaSpmdShardToFullShape<T> xlaSpmdShardToFullShape(Oper
    * @return a constant tensor initialized with zeros
    * @throws IllegalArgumentException if the tensor type or shape cannot be initialized with zeros.
    */
-  public <T extends TType, U extends TNumber> Zeros<T> zeros(Operand<U> dims, DataType<T> type) {
+  public <T extends TType, U extends TNumber> Zeros<T> zeros(Operand<U> dims, Class<T> type) {
     return Zeros.create(scope, dims, type);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/QuantizationOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/QuantizationOps.java
index aec0d667c65..045ed3d4b03 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/QuantizationOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/QuantizationOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.quantization.Dequantize;
 import org.tensorflow.op.quantization.FakeQuantWithMinMaxArgs;
@@ -173,7 +172,7 @@ public <T extends TType> Dequantize<TFloat32> dequantize(Operand<T> input,
    * @return a new instance of Dequantize
    */
   public <U extends TNumber, T extends TType> Dequantize<U> dequantize(Operand<T> input,
-      Operand<TFloat32> minRange, Operand<TFloat32> maxRange, DataType<U> dtype,
+      Operand<TFloat32> minRange, Operand<TFloat32> maxRange, Class<U> dtype,
       Dequantize.Options... options) {
     return Dequantize.create(scope, input, minRange, maxRange, dtype, options);
   }
@@ -504,7 +503,7 @@ public FakeQuantWithMinMaxVarsPerChannelGradient fakeQuantWithMinMaxVarsPerChann
    * @return a new instance of Quantize
    */
   public <T extends TType> Quantize<T> quantize(Operand<TFloat32> input, Operand<TFloat32> minRange,
-      Operand<TFloat32> maxRange, DataType<T> T, Quantize.Options... options) {
+      Operand<TFloat32> maxRange, Class<T> T, Quantize.Options... options) {
     return Quantize.create(scope, input, minRange, maxRange, T, options);
   }
 
@@ -562,8 +561,7 @@ public <T extends TNumber> QuantizeAndDequantize<T> quantizeAndDequantize(Operan
    * @return a new instance of QuantizeDownAndShrinkRange
    */
   public <U extends TType, T extends TType> QuantizeDownAndShrinkRange<U> quantizeDownAndShrinkRange(
-      Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax,
-      DataType<U> outType) {
+      Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, Class<U> outType) {
     return QuantizeDownAndShrinkRange.create(scope, input, inputMin, inputMax, outType);
   }
 
@@ -625,7 +623,7 @@ public <T extends TType> RequantizationRange requantizationRange(Operand<T> inpu
    */
   public <U extends TType, T extends TType> Requantize<U> requantize(Operand<T> input,
       Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, Operand<TFloat32> requestedOutputMin,
-      Operand<TFloat32> requestedOutputMax, DataType<U> outType) {
+      Operand<TFloat32> requestedOutputMax, Class<U> outType) {
     return Requantize.create(scope, input, inputMin, inputMax, requestedOutputMin, requestedOutputMax, outType);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/RandomOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/RandomOps.java
index 071e77c7a70..5fe5b6268fe 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/RandomOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/RandomOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.random.AllCandidateSampler;
 import org.tensorflow.op.random.LogUniformCandidateSampler;
@@ -141,7 +140,7 @@ public <T extends TNumber> Multinomial<TInt64> multinomial(Operand<T> logits,
    * @return a new instance of Multinomial
    */
   public <U extends TNumber, T extends TNumber> Multinomial<U> multinomial(Operand<T> logits,
-      Operand<TInt32> numSamples, DataType<U> outputDtype, Multinomial.Options... options) {
+      Operand<TInt32> numSamples, Class<U> outputDtype, Multinomial.Options... options) {
     return Multinomial.create(scope, logits, numSamples, outputDtype, options);
   }
 
@@ -236,7 +235,7 @@ public <T extends TNumber, U extends TNumber> RandomPoisson<TInt64> randomPoisso
    * @return a new instance of RandomPoisson
    */
   public <V extends TNumber, T extends TNumber, U extends TNumber> RandomPoisson<V> randomPoisson(
-      Operand<T> shape, Operand<U> rate, DataType<V> dtype, RandomPoisson.Options... options) {
+      Operand<T> shape, Operand<U> rate, Class<V> dtype, RandomPoisson.Options... options) {
     return RandomPoisson.create(scope, shape, rate, dtype, options);
   }
 
@@ -274,7 +273,7 @@ public <T extends TType> RandomShuffle<T> randomShuffle(Operand<T> value,
    * @return a new instance of RandomStandardNormal
    */
   public <U extends TNumber, T extends TNumber> RandomStandardNormal<U> randomStandardNormal(
-      Operand<T> shape, DataType<U> dtype, RandomStandardNormal.Options... options) {
+      Operand<T> shape, Class<U> dtype, RandomStandardNormal.Options... options) {
     return RandomStandardNormal.create(scope, shape, dtype, options);
   }
 
@@ -291,7 +290,7 @@ public <U extends TNumber, T extends TNumber> RandomStandardNormal<U> randomStan
    * @return a new instance of RandomUniform
    */
   public <U extends TNumber, T extends TNumber> RandomUniform<U> randomUniform(Operand<T> shape,
-      DataType<U> dtype, RandomUniform.Options... options) {
+      Class<U> dtype, RandomUniform.Options... options) {
     return RandomUniform.create(scope, shape, dtype, options);
   }
 
@@ -358,7 +357,7 @@ public <T extends TNumber, U extends TNumber> StatefulRandomBinomial<TInt64> sta
    */
   public <V extends TNumber, T extends TNumber, U extends TNumber> StatefulRandomBinomial<V> statefulRandomBinomial(
       Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Operand<U> counts,
-      Operand<U> probs, DataType<V> dtype) {
+      Operand<U> probs, Class<V> dtype) {
     return StatefulRandomBinomial.create(scope, resource, algorithm, shape, counts, probs, dtype);
   }
 
@@ -391,7 +390,7 @@ public <T extends TType> StatefulStandardNormal<TFloat32> statefulStandardNormal
    * @return a new instance of StatefulStandardNormal
    */
   public <U extends TType, T extends TType> StatefulStandardNormal<U> statefulStandardNormal(
-      Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, DataType<U> dtype) {
+      Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Class<U> dtype) {
     return StatefulStandardNormal.create(scope, resource, algorithm, shape, dtype);
   }
 
@@ -422,7 +421,7 @@ public <T extends TNumber, U extends TNumber> StatelessMultinomial<TInt64> state
    * @return a new instance of StatelessMultinomial
    */
   public <V extends TNumber, T extends TNumber, U extends TNumber> StatelessMultinomial<V> statelessMultinomial(
-      Operand<T> logits, Operand<TInt32> numSamples, Operand<U> seed, DataType<V> outputDtype) {
+      Operand<T> logits, Operand<TInt32> numSamples, Operand<U> seed, Class<V> outputDtype) {
     return StatelessMultinomial.create(scope, logits, numSamples, seed, outputDtype);
   }
 
@@ -457,7 +456,7 @@ public <T extends TNumber, U extends TNumber> StatelessRandomNormal<TFloat32> st
    * @return a new instance of StatelessRandomNormal
    */
   public <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomNormal<V> statelessRandomNormal(
-      Operand<T> shape, Operand<U> seed, DataType<V> dtype) {
+      Operand<T> shape, Operand<U> seed, Class<V> dtype) {
     return StatelessRandomNormal.create(scope, shape, seed, dtype);
   }
 
@@ -494,7 +493,7 @@ public <T extends TNumber, U extends TNumber> StatelessRandomUniform<TFloat32> s
    * @return a new instance of StatelessRandomUniform
    */
   public <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomUniform<V> statelessRandomUniform(
-      Operand<T> shape, Operand<U> seed, DataType<V> dtype) {
+      Operand<T> shape, Operand<U> seed, Class<V> dtype) {
     return StatelessRandomUniform.create(scope, shape, seed, dtype);
   }
 
@@ -533,7 +532,7 @@ public <T extends TNumber, U extends TNumber> StatelessTruncatedNormal<TFloat32>
    * @return a new instance of StatelessTruncatedNormal
    */
   public <V extends TNumber, T extends TNumber, U extends TNumber> StatelessTruncatedNormal<V> statelessTruncatedNormal(
-      Operand<T> shape, Operand<U> seed, DataType<V> dtype) {
+      Operand<T> shape, Operand<U> seed, Class<V> dtype) {
     return StatelessTruncatedNormal.create(scope, shape, seed, dtype);
   }
 
@@ -551,7 +550,7 @@ public <V extends TNumber, T extends TNumber, U extends TNumber> StatelessTrunca
    * @return a new instance of TruncatedNormal
    */
   public <U extends TNumber, T extends TNumber> TruncatedNormal<U> truncatedNormal(Operand<T> shape,
-      DataType<U> dtype, TruncatedNormal.Options... options) {
+      Class<U> dtype, TruncatedNormal.Options... options) {
     return TruncatedNormal.create(scope, shape, dtype, options);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ShapeOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ShapeOps.java
index 81c692571f1..3683805eb2e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ShapeOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/ShapeOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.core.Shape;
 import org.tensorflow.op.core.Shapes;
@@ -116,7 +115,7 @@ public Operand<TInt32> flatten(Shape<TInt32> shape) {
    * @return the reshaped operand
    */
   public <T extends TType, U extends TNumber> Operand<T> flatten(Operand<T> operand,
-      DataType<U> dType) {
+      Class<U> dType) {
     return Shapes.flatten(scope, operand, dType);
   }
 
@@ -129,7 +128,7 @@ public <T extends TType, U extends TNumber> Operand<T> flatten(Operand<T> operan
    * @param dType the shape datatype
    * @return the flattened shape
    */
-  public <U extends TNumber> Operand<U> flatten(Shape<U> shape, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> flatten(Shape<U> shape, Class<U> dType) {
     return Shapes.flatten(scope, shape, dType);
   }
 
@@ -153,7 +152,7 @@ public Operand<TInt32> head(Shape<TInt32> shape) {
    * @param <U> the shape datatype.
    * @return a 1-dimensional Operand containing the Shape's first dimension
    */
-  public <U extends TNumber> Operand<U> head(Shape<U> shape, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> head(Shape<U> shape, Class<U> dType) {
     return Shapes.head(scope, shape, dType);
   }
 
@@ -177,7 +176,7 @@ public Operand<TInt32> numDimensions(Shape<TInt32> shape) {
    * @param dType the shape datatype
    * @return the number of dimensions
    */
-  public <U extends TNumber> Operand<U> numDimensions(Shape<U> shape, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> numDimensions(Shape<U> shape, Class<U> dType) {
     return Shapes.numDimensions(scope, shape, dType);
   }
 
@@ -262,7 +261,7 @@ public Operand<TInt32> reduceDims(Shape<TInt32> shape, Operand<TInt32> axis) {
    * @return the reshaped operand
    */
   public <T extends TType, U extends TNumber> Operand<T> reduceDims(Operand<T> operand,
-      Operand<U> axis, DataType<U> dType) {
+      Operand<U> axis, Class<U> dType) {
     return Shapes.reduceDims(scope, operand, axis, dType);
   }
 
@@ -277,7 +276,7 @@ public <T extends TType, U extends TNumber> Operand<T> reduceDims(Operand<T> ope
    * @return the reduced shape
    */
   public <U extends TNumber> Operand<U> reduceDims(Shape<U> shape, Operand<U> axis,
-      DataType<U> dType) {
+      Class<U> dType) {
     return Shapes.reduceDims(scope, shape, axis, dType);
   }
 
@@ -313,7 +312,7 @@ public <T extends TType> Operand<TInt32> size(Operand<T> input, Operand<TInt32>
    * @param dType the shape datatype
    * @return the size
    */
-  public <U extends TNumber> Operand<U> size(Shape<U> shape, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> size(Shape<U> shape, Class<U> dType) {
     return Shapes.size(scope, shape, dType);
   }
 
@@ -340,7 +339,7 @@ public Operand<TInt32> size(Shape<TInt32> shape, Operand<TInt32> dim) {
    * @return the size of the specified dimension
    */
   public <T extends TType, U extends TNumber> Operand<U> size(Operand<T> input, Operand<U> dim,
-      DataType<U> dType) {
+      Class<U> dType) {
     return Shapes.size(scope, input, dim, dType);
   }
 
@@ -354,7 +353,7 @@ public <T extends TType, U extends TNumber> Operand<U> size(Operand<T> input, Op
    * @param dType the shape datatype
    * @return the size of the specified dimension
    */
-  public <U extends TNumber> Operand<U> size(Shape<U> shape, Operand<U> dim, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> size(Shape<U> shape, Operand<U> dim, Class<U> dType) {
     return Shapes.size(scope, shape, dim, dType);
   }
 
@@ -378,7 +377,7 @@ public Operand<TInt32> squeeze(Shape<TInt32> shape) {
    * @param dType the shape datatype.
    * @return the squeezed shape
    */
-  public <U extends TNumber> Operand<U> squeeze(Shape<U> shape, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> squeeze(Shape<U> shape, Class<U> dType) {
     return Shapes.squeeze(scope, shape, dType);
   }
 
@@ -406,7 +405,7 @@ public Operand<TInt32> tail(Shape<TInt32> shape) {
    * @return a 1-dimensional Operand that contains the dimension matching the last dimension of the
    *      Shape
    */
-  public <U extends TNumber> Operand<U> tail(Shape<U> shape, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> tail(Shape<U> shape, Class<U> dType) {
     return Shapes.tail(scope, shape, dType);
   }
 
@@ -436,7 +435,7 @@ public Operand<TInt32> take(Shape<TInt32> shape, Operand<TInt32> n) {
    * @return a 1-dimensional operand with the dimensions matching * the first n dimensions of the
    *      shape
    */
-  public <U extends TNumber> Operand<U> take(Shape<U> shape, Operand<U> n, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> take(Shape<U> shape, Operand<U> n, Class<U> dType) {
     return Shapes.take(scope, shape, n, dType);
   }
 
@@ -466,7 +465,7 @@ public <U extends TNumber> Operand<TInt32> takeLast(Shape<TInt32> shape, Operand
    * @return a 1-dimensional operand containing the dimensions matching the last n dimensions of the
    *      shape
    */
-  public <U extends TNumber> Operand<U> takeLast(Shape<U> shape, Operand<U> n, DataType<U> dType) {
+  public <U extends TNumber> Operand<U> takeLast(Shape<U> shape, Operand<U> n, Class<U> dType) {
     return Shapes.takeLast(scope, shape, n, dType);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SignalOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SignalOps.java
index f4ec7bdb48d..1466dcd8de4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SignalOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SignalOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.signal.BatchFft;
 import org.tensorflow.op.signal.BatchFft2d;
@@ -242,7 +241,7 @@ public <T extends TType> Irfft<TFloat32> irfft(Operand<T> input, Operand<TInt32>
    * @return a new instance of Irfft
    */
   public <U extends TNumber, T extends TType> Irfft<U> irfft(Operand<T> input,
-      Operand<TInt32> fftLength, DataType<U> Treal) {
+      Operand<TInt32> fftLength, Class<U> Treal) {
     return Irfft.create(scope, input, fftLength, Treal);
   }
 
@@ -298,7 +297,7 @@ public <T extends TType> Irfft2d<TFloat32> irfft2d(Operand<T> input, Operand<TIn
    * @return a new instance of Irfft2d
    */
   public <U extends TNumber, T extends TType> Irfft2d<U> irfft2d(Operand<T> input,
-      Operand<TInt32> fftLength, DataType<U> Treal) {
+      Operand<TInt32> fftLength, Class<U> Treal) {
     return Irfft2d.create(scope, input, fftLength, Treal);
   }
 
@@ -354,7 +353,7 @@ public <T extends TType> Irfft3d<TFloat32> irfft3d(Operand<T> input, Operand<TIn
    * @return a new instance of Irfft3d
    */
   public <U extends TNumber, T extends TType> Irfft3d<U> irfft3d(Operand<T> input,
-      Operand<TInt32> fftLength, DataType<U> Treal) {
+      Operand<TInt32> fftLength, Class<U> Treal) {
     return Irfft3d.create(scope, input, fftLength, Treal);
   }
 
@@ -379,7 +378,7 @@ public <U extends TNumber, T extends TType> Irfft3d<U> irfft3d(Operand<T> input,
    * @return a new instance of Rfft
    */
   public <U extends TType, T extends TNumber> Rfft<U> rfft(Operand<T> input,
-      Operand<TInt32> fftLength, DataType<U> Tcomplex) {
+      Operand<TInt32> fftLength, Class<U> Tcomplex) {
     return Rfft.create(scope, input, fftLength, Tcomplex);
   }
 
@@ -405,7 +404,7 @@ public <U extends TType, T extends TNumber> Rfft<U> rfft(Operand<T> input,
    * @return a new instance of Rfft2d
    */
   public <U extends TType, T extends TNumber> Rfft2d<U> rfft2d(Operand<T> input,
-      Operand<TInt32> fftLength, DataType<U> Tcomplex) {
+      Operand<TInt32> fftLength, Class<U> Tcomplex) {
     return Rfft2d.create(scope, input, fftLength, Tcomplex);
   }
 
@@ -431,7 +430,7 @@ public <U extends TType, T extends TNumber> Rfft2d<U> rfft2d(Operand<T> input,
    * @return a new instance of Rfft3d
    */
   public <U extends TType, T extends TNumber> Rfft3d<U> rfft3d(Operand<T> input,
-      Operand<TInt32> fftLength, DataType<U> Tcomplex) {
+      Operand<TInt32> fftLength, Class<U> Tcomplex) {
     return Rfft3d.create(scope, input, fftLength, Tcomplex);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SparseOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SparseOps.java
index 42cdf9569d9..d7690dacdb4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SparseOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/SparseOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.sparse.AddManySparseToTensorsMap;
@@ -269,7 +268,7 @@ public <T extends TType> DenseToSparseSetOperation<T> denseToSparseSetOperation(
    * @return a new instance of DeserializeSparse
    */
   public <U extends TType, T extends TType> DeserializeSparse<U> deserializeSparse(
-      Operand<T> serializedSparse, DataType<U> dtype) {
+      Operand<T> serializedSparse, Class<U> dtype) {
     return DeserializeSparse.create(scope, serializedSparse, dtype);
   }
 
@@ -315,7 +314,7 @@ public <T extends TType> SparseAccumulatorApplyGradient sparseAccumulatorApplyGr
    * @return a new instance of SparseAccumulatorTakeGradient
    */
   public <T extends TType> SparseAccumulatorTakeGradient<T> sparseAccumulatorTakeGradient(
-      Operand<TString> handle, Operand<TInt32> numRequired, DataType<T> dtype) {
+      Operand<TString> handle, Operand<TInt32> numRequired, Class<T> dtype) {
     return SparseAccumulatorTakeGradient.create(scope, handle, numRequired, dtype);
   }
 
@@ -476,8 +475,8 @@ public <T extends TType> SparseConcat<T> sparseConcat(Iterable<Operand<TInt64>>
    * @param options carries optional attributes values
    * @return a new instance of SparseConditionalAccumulator
    */
-  public <T extends TType> SparseConditionalAccumulator sparseConditionalAccumulator(
-      DataType<T> dtype, Shape shape, SparseConditionalAccumulator.Options... options) {
+  public <T extends TType> SparseConditionalAccumulator sparseConditionalAccumulator(Class<T> dtype,
+      Shape shape, SparseConditionalAccumulator.Options... options) {
     return SparseConditionalAccumulator.create(scope, dtype, shape, options);
   }
 
@@ -1493,7 +1492,7 @@ public <T extends TType> SparseToSparseSetOperation<T> sparseToSparseSetOperatio
    * @return a new instance of TakeManySparseFromTensorsMap
    */
   public <T extends TType> TakeManySparseFromTensorsMap<T> takeManySparseFromTensorsMap(
-      Operand<TInt64> sparseHandles, DataType<T> dtype,
+      Operand<TInt64> sparseHandles, Class<T> dtype,
       TakeManySparseFromTensorsMap.Options... options) {
     return TakeManySparseFromTensorsMap.create(scope, sparseHandles, dtype, options);
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/StringsOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/StringsOps.java
index f6491843332..a09884817f8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/StringsOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/StringsOps.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.strings.Join;
 import org.tensorflow.op.strings.Lower;
@@ -483,8 +482,7 @@ public ToNumber<TFloat32> toNumber(Operand<TString> stringTensor) {
    * @param outType The numeric type to interpret each string in `string_tensor` as.
    * @return a new instance of ToNumber
    */
-  public <T extends TNumber> ToNumber<T> toNumber(Operand<TString> stringTensor,
-      DataType<T> outType) {
+  public <T extends TNumber> ToNumber<T> toNumber(Operand<TString> stringTensor, Class<T> outType) {
     return ToNumber.create(scope, stringTensor, outType);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/TrainOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/TrainOps.java
index 2c5d8752136..1e253ac9956 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/TrainOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/TrainOps.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.train.AccumulatorApplyGradient;
@@ -159,7 +158,7 @@ public AccumulatorSetGlobalStep accumulatorSetGlobalStep(Operand<TString> handle
    * @return a new instance of AccumulatorTakeGradient
    */
   public <T extends TType> AccumulatorTakeGradient<T> accumulatorTakeGradient(
-      Operand<TString> handle, Operand<TInt32> numRequired, DataType<T> dtype) {
+      Operand<TString> handle, Operand<TInt32> numRequired, Class<T> dtype) {
     return AccumulatorTakeGradient.create(scope, handle, numRequired, dtype);
   }
 
@@ -542,7 +541,7 @@ public <T extends TType> BatchMatMul<T> batchMatMul(Operand<T> x, Operand<T> y,
    * @param options carries optional attributes values
    * @return a new instance of ConditionalAccumulator
    */
-  public <T extends TType> ConditionalAccumulator conditionalAccumulator(DataType<T> dtype,
+  public <T extends TType> ConditionalAccumulator conditionalAccumulator(Class<T> dtype,
       Shape shape, ConditionalAccumulator.Options... options) {
     return ConditionalAccumulator.create(scope, dtype, shape, options);
   }
@@ -1295,7 +1294,7 @@ public <T extends TType, U extends TNumber> ResourceSparseApplyRmsProp resourceS
    * @return a new instance of Restore
    */
   public Restore restore(Operand<TString> prefix, Operand<TString> tensorNames,
-      Operand<TString> shapeAndSlices, List<DataType<?>> dtypes) {
+      Operand<TString> shapeAndSlices, List<Class<? extends TType>> dtypes) {
     return Restore.create(scope, prefix, tensorNames, shapeAndSlices, dtypes);
   }
 
@@ -1321,7 +1320,7 @@ public Restore restore(Operand<TString> prefix, Operand<TString> tensorNames,
    * @return a new instance of RestoreSlice
    */
   public <T extends TType> RestoreSlice<T> restoreSlice(Operand<TString> filePattern,
-      Operand<TString> tensorName, Operand<TString> shapeAndSlice, DataType<T> dt,
+      Operand<TString> tensorName, Operand<TString> shapeAndSlice, Class<T> dt,
       RestoreSlice.Options... options) {
     return RestoreSlice.create(scope, filePattern, tensorName, shapeAndSlice, dt, options);
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/XlaOps.java b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/XlaOps.java
index 535972d4883..6c16e3cb7ff 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/XlaOps.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/XlaOps.java
@@ -17,7 +17,6 @@
 //
 package org.tensorflow.op;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.xla.BroadcastHelper;
@@ -278,7 +277,7 @@ public <T extends TType, U extends TNumber> Pad<T> pad(Operand<T> input, Operand
    * @param shape The shape of the tensor.
    * @return a new instance of Recv
    */
-  public <T extends TType> Recv<T> recv(DataType<T> dtype, String tensorName, Shape shape) {
+  public <T extends TType> Recv<T> recv(Class<T> dtype, String tensorName, Shape shape) {
     return Recv.create(scope, dtype, tensorName, shape);
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseAnd.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseAnd.java
index e199ff2201f..2e3d0cb2949 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseAnd.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseAnd.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Elementwise computes the bitwise AND of `x` and `y`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseOr.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseOr.java
index 264c2bc340b..abeb509f5c7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseOr.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseOr.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Elementwise computes the bitwise OR of `x` and `y`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseXor.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseXor.java
index 1d8f668c175..4339d2d1df9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseXor.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/BitwiseXor.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Elementwise computes the bitwise XOR of `x` and `y`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/Invert.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/Invert.java
index 9f8bdfd56d8..589f0047356 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/Invert.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/Invert.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Invert (flip) each bit of supported types; for example, type `uint8` value 01010101 becomes 10101010.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/LeftShift.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/LeftShift.java
index f7a47534d81..69c818873e9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/LeftShift.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/LeftShift.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Elementwise computes the bitwise left-shift of `x` and `y`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/RightShift.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/RightShift.java
index 99c5fe5766e..8106172ee4f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/RightShift.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/bitwise/RightShift.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Elementwise computes the bitwise right-shift of `x` and `y`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/AllReduce.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/AllReduce.java
index d58bc1357f4..0ef37a5a9d2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/AllReduce.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/AllReduce.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Mutually reduces multiple tensors of identical type and shape.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/BroadcastRecv.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/BroadcastRecv.java
index ab938852275..8603acc0e05 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/BroadcastRecv.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/collective/BroadcastRecv.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.collective;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -77,7 +76,7 @@ private Options() {
    * @return a new instance of BroadcastRecv
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> BroadcastRecv<T> create(Scope scope, DataType<T> T, Long groupSize, Long groupKey, Long instanceKey, Shape shape, Options... options) {
+  public static <T extends TType> BroadcastRecv<T> create(Scope scope, Class<T> T, Long groupSize, Long groupKey, Long instanceKey, Shape shape, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("CollectiveBcastRecv", scope.makeOpName("BroadcastRecv"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("T", T);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/All.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/All.java
index 909427d1a57..8de82899104 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/All.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/All.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the "logical and" of elements across dimensions of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Any.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Any.java
index 0316e5e1a94..f5e5f3753e2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Any.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Any.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the "logical or" of elements across dimensions of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Barrier.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Barrier.java
index b9c5c84083f..dc3f8b781bc 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Barrier.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Barrier.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,6 +28,7 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 /**
  * Defines a barrier that persists across different graph executions.
@@ -105,10 +105,10 @@ private Options() {
    * @return a new instance of Barrier
    */
   @Endpoint(describeByClass = true)
-  public static Barrier create(Scope scope, List<DataType<?>> componentTypes, Options... options) {
+  public static Barrier create(Scope scope, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Barrier", scope.makeOpName("Barrier"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BarrierTakeMany.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BarrierTakeMany.java
index 6c391fab5fa..9032635571f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BarrierTakeMany.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BarrierTakeMany.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -31,6 +30,7 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 /**
  * Takes the given number of completed elements from a barrier.
@@ -98,12 +98,12 @@ private Options() {
    * @return a new instance of BarrierTakeMany
    */
   @Endpoint(describeByClass = true)
-  public static BarrierTakeMany create(Scope scope, Operand<TString> handle, Operand<TInt32> numElements, List<DataType<?>> componentTypes, Options... options) {
+  public static BarrierTakeMany create(Scope scope, Operand<TString> handle, Operand<TInt32> numElements, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("BarrierTakeMany", scope.makeOpName("BarrierTakeMany"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(numElements.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bitcast.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bitcast.java
index b01c8598ae6..65d39833a78 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bitcast.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bitcast.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -96,7 +95,7 @@ public final class Bitcast<U extends TType> extends RawOp implements Operand<U>
    * @return a new instance of Bitcast
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> Bitcast<U> create(Scope scope, Operand<T> input, DataType<U> type) {
+  public static <U extends TType, T extends TType> Bitcast<U> create(Scope scope, Operand<T> input, Class<U> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("Bitcast", scope.makeOpName("Bitcast"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastDynamicShape.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastDynamicShape.java
index 3027e8234a4..df29691cac8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastDynamicShape.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastDynamicShape.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Return the shape of s0 op s1 with broadcast.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastGradientArgs.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastGradientArgs.java
index 2d95c71086e..dffa39e4bf1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastGradientArgs.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/BroadcastGradientArgs.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Return the reduction indices for computing gradients of s0 op s1 with broadcast.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bucketize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bucketize.java
index 84e9b454d0d..05b43dc7679 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bucketize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Bucketize.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Bucketizes 'input' based on 'boundaries'.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CollectiveGather.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CollectiveGather.java
index bd9de3ebb33..d1f4e84f146 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CollectiveGather.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CollectiveGather.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Mutually accumulates multiple tensors of identical type and shape.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CountUpTo.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CountUpTo.java
index 2884783695d..811a955d2fd 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CountUpTo.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/CountUpTo.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Increments 'ref' until it reaches 'limit'.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/DecodeProto.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/DecodeProto.java
index 28af55f8465..40afa08fc97 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/DecodeProto.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/DecodeProto.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -30,6 +29,7 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 /**
  * The op extracts fields from a serialized protocol buffers message into tensors.
@@ -135,7 +135,7 @@ private Options() {
    * @return a new instance of DecodeProto
    */
   @Endpoint(describeByClass = true)
-  public static DecodeProto create(Scope scope, Operand<TString> bytes, String messageType, List<String> fieldNames, List<DataType<?>> outputTypes, Options... options) {
+  public static DecodeProto create(Scope scope, Operand<TString> bytes, String messageType, List<String> fieldNames, List<Class<? extends TType>> outputTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("DecodeProtoV2", scope.makeOpName("DecodeProto"));
     opBuilder.addInput(bytes.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -145,7 +145,7 @@ public static DecodeProto create(Scope scope, Operand<TString> bytes, String mes
       fieldNamesArray[i] = fieldNames.get(i);
     }
     opBuilder.setAttr("field_names", fieldNamesArray);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/DummySeedGenerator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/DummySeedGenerator.java
deleted file mode 100644
index b9cf2c36d09..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/DummySeedGenerator.java
+++ /dev/null
@@ -1,66 +0,0 @@
-/* Copyright 2018 The TensorFlow Authors. All Rights Reserved.
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-=======================================================================*/
-
-// This class has been generated, DO NOT EDIT!
-
-package org.tensorflow.op.core;
-
-import org.tensorflow.Operand;
-import org.tensorflow.Operation;
-import org.tensorflow.OperationBuilder;
-import org.tensorflow.Output;
-import org.tensorflow.op.RawOp;
-import org.tensorflow.op.Scope;
-import org.tensorflow.op.annotation.Endpoint;
-import org.tensorflow.op.annotation.Operator;
-import org.tensorflow.types.family.TType;
-
-/**
- */
-public final class DummySeedGenerator extends RawOp implements Operand<TType> {
-  
-  /**
-   * Factory method to create a class wrapping a new DummySeedGenerator operation.
-   * 
-   * @param scope current scope
-   * @return a new instance of DummySeedGenerator
-   */
-  @Endpoint(describeByClass = true)
-  public static DummySeedGenerator create(Scope scope) {
-    OperationBuilder opBuilder = scope.env().opBuilder("DummySeedGenerator", scope.makeOpName("DummySeedGenerator"));
-    opBuilder = scope.applyControlDependencies(opBuilder);
-    return new DummySeedGenerator(opBuilder.build());
-  }
-  
-  /**
-   */
-  public Output<?> handle() {
-    return handle;
-  }
-  
-  @Override
-  @SuppressWarnings("unchecked")
-  public Output<TType> asOutput() {
-    return (Output<TType>) handle;
-  }
-  
-  private Output<?> handle;
-  
-  private DummySeedGenerator(Operation operation) {
-    super(operation);
-    int outputIdx = 0;
-    handle = operation.output(outputIdx++);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Empty.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Empty.java
index c8305349d37..9330b2dfee3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Empty.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Empty.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -68,7 +67,7 @@ private Options() {
    * @return a new instance of Empty
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> Empty<T> create(Scope scope, Operand<TInt32> shape, DataType<T> dtype, Options... options) {
+  public static <T extends TType> Empty<T> create(Scope scope, Operand<TInt32> shape, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Empty", scope.makeOpName("Empty"));
     opBuilder.addInput(shape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/EmptyTensorList.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/EmptyTensorList.java
index 619fb90657f..ecd16df5097 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/EmptyTensorList.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/EmptyTensorList.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -53,7 +52,7 @@ public final class EmptyTensorList extends RawOp implements Operand<TType> {
    * @return a new instance of EmptyTensorList
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber, U extends TType> EmptyTensorList create(Scope scope, Operand<T> elementShape, Operand<TInt32> maxNumElements, DataType<U> elementDtype) {
+  public static <T extends TNumber, U extends TType> EmptyTensorList create(Scope scope, Operand<T> elementShape, Operand<TInt32> maxNumElements, Class<U> elementDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("EmptyTensorList", scope.makeOpName("EmptyTensorList"));
     opBuilder.addInput(elementShape.asOutput());
     opBuilder.addInput(maxNumElements.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ExtractVolumePatches.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ExtractVolumePatches.java
index 0b9bcd78e20..487289a8a48 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ExtractVolumePatches.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ExtractVolumePatches.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Extract `patches` from `input` and put them in the "depth" output dimension. 3D extension of `extract_image_patches`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/GetSessionTensor.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/GetSessionTensor.java
index 93ba5af508c..d5767515e56 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/GetSessionTensor.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/GetSessionTensor.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,7 +45,7 @@ public final class GetSessionTensor<T extends TType> extends RawOp implements Op
    * @return a new instance of GetSessionTensor
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> GetSessionTensor<T> create(Scope scope, Operand<TString> handle, DataType<T> dtype) {
+  public static <T extends TType> GetSessionTensor<T> create(Scope scope, Operand<TString> handle, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("GetSessionTensor", scope.makeOpName("GetSessionTensor"));
     opBuilder.addInput(handle.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HashTable.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HashTable.java
index 87d9cab4c3f..aea656de1ec 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HashTable.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HashTable.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -88,7 +87,7 @@ private Options() {
    * @return a new instance of HashTable
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TType> HashTable create(Scope scope, DataType<T> keyDtype, DataType<U> valueDtype, Options... options) {
+  public static <T extends TType, U extends TType> HashTable create(Scope scope, Class<T> keyDtype, Class<U> valueDtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("HashTableV2", scope.makeOpName("HashTable"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("key_dtype", keyDtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HistogramFixedWidth.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HistogramFixedWidth.java
index da1c1a7b713..1565dd2a8d0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HistogramFixedWidth.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/HistogramFixedWidth.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Return histogram of values.
@@ -67,7 +65,7 @@ public final class HistogramFixedWidth<U extends TNumber> extends RawOp implemen
    * @return a new instance of HistogramFixedWidth
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TNumber> HistogramFixedWidth<U> create(Scope scope, Operand<T> values, Operand<T> valueRange, Operand<TInt32> nbins, DataType<U> dtype) {
+  public static <U extends TNumber, T extends TNumber> HistogramFixedWidth<U> create(Scope scope, Operand<T> values, Operand<T> valueRange, Operand<TInt32> nbins, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("HistogramFixedWidth", scope.makeOpName("HistogramFixedWidth"));
     opBuilder.addInput(values.asOutput());
     opBuilder.addInput(valueRange.asOutput());
@@ -90,7 +88,7 @@ public static <U extends TNumber, T extends TNumber> HistogramFixedWidth<U> crea
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber> HistogramFixedWidth<TInt32> create(Scope scope, Operand<T> values, Operand<T> valueRange, Operand<TInt32> nbins) {
-    return create(scope, values, valueRange, nbins, TInt32.DTYPE);
+    return create(scope, values, valueRange, nbins, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ImmutableConst.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ImmutableConst.java
index ecbc3154498..3d2ac9625da 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ImmutableConst.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ImmutableConst.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -50,7 +49,7 @@ public final class ImmutableConst<T extends TType> extends RawOp implements Oper
    * @return a new instance of ImmutableConst
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> ImmutableConst<T> create(Scope scope, DataType<T> dtype, Shape shape, String memoryRegionName) {
+  public static <T extends TType> ImmutableConst<T> create(Scope scope, Class<T> dtype, Shape shape, String memoryRegionName) {
     OperationBuilder opBuilder = scope.env().opBuilder("ImmutableConst", scope.makeOpName("ImmutableConst"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LinSpace.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LinSpace.java
index 04f3959820a..fbfa0e69881 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LinSpace.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LinSpace.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Generates values in an interval.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LookupTableExport.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LookupTableExport.java
index 6bebbb35895..527e70e7fc0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LookupTableExport.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LookupTableExport.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,7 +46,7 @@ public final class LookupTableExport<T extends TType, U extends TType> extends R
    * @return a new instance of LookupTableExport
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TType> LookupTableExport<T, U> create(Scope scope, Operand<?> tableHandle, DataType<T> Tkeys, DataType<U> Tvalues) {
+  public static <T extends TType, U extends TType> LookupTableExport<T, U> create(Scope scope, Operand<?> tableHandle, Class<T> Tkeys, Class<U> Tvalues) {
     OperationBuilder opBuilder = scope.env().opBuilder("LookupTableExportV2", scope.makeOpName("LookupTableExport"));
     opBuilder.addInput(tableHandle.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LowerBound.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LowerBound.java
index 6f0f5158cca..7ec77e7df0b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LowerBound.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/LowerBound.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -66,7 +65,7 @@ public final class LowerBound<U extends TNumber> extends RawOp implements Operan
    * @return a new instance of LowerBound
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> LowerBound<U> create(Scope scope, Operand<T> sortedInputs, Operand<T> values, DataType<U> outType) {
+  public static <U extends TNumber, T extends TType> LowerBound<U> create(Scope scope, Operand<T> sortedInputs, Operand<T> values, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("LowerBound", scope.makeOpName("LowerBound"));
     opBuilder.addInput(sortedInputs.asOutput());
     opBuilder.addInput(values.asOutput());
@@ -86,7 +85,7 @@ public static <U extends TNumber, T extends TType> LowerBound<U> create(Scope sc
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> LowerBound<TInt32> create(Scope scope, Operand<T> sortedInputs, Operand<T> values) {
-    return create(scope, sortedInputs, values, TInt32.DTYPE);
+    return create(scope, sortedInputs, values, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapClear.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapClear.java
index bad1e90554f..3c3dca46cc6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapClear.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapClear.java
@@ -18,13 +18,13 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
 import org.tensorflow.op.RawOp;
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op removes all elements in the underlying container.
@@ -87,10 +87,10 @@ private Options() {
    * @return a new instance of MapClear
    */
   @Endpoint(describeByClass = true)
-  public static MapClear create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static MapClear create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MapClear", scope.makeOpName("MapClear"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapIncompleteSize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapIncompleteSize.java
index 19e9e87a08a..76b1513c6bd 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapIncompleteSize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapIncompleteSize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,6 +27,7 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op returns the number of incomplete elements in the underlying container.
@@ -90,10 +90,10 @@ private Options() {
    * @return a new instance of MapIncompleteSize
    */
   @Endpoint(describeByClass = true)
-  public static MapIncompleteSize create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static MapIncompleteSize create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MapIncompleteSize", scope.makeOpName("MapIncompleteSize"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapPeek.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapPeek.java
index 1925ca680ea..46dccfc6a3f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapPeek.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapPeek.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -99,12 +98,12 @@ private Options() {
    * @return a new instance of MapPeek
    */
   @Endpoint(describeByClass = true)
-  public static MapPeek create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<DataType<?>> dtypes, Options... options) {
+  public static MapPeek create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MapPeek", scope.makeOpName("MapPeek"));
     opBuilder.addInput(key.asOutput());
     opBuilder.addInput(indices.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapSize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapSize.java
index 7f4eea906f5..600b9473da9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapSize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapSize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,6 +27,7 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op returns the number of elements in the underlying container.
@@ -90,10 +90,10 @@ private Options() {
    * @return a new instance of MapSize
    */
   @Endpoint(describeByClass = true)
-  public static MapSize create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static MapSize create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MapSize", scope.makeOpName("MapSize"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapStage.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapStage.java
index 9291b32d53b..eb75d99ba5c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapStage.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapStage.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,6 +28,7 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TType;
 
 /**
  * Stage (key, values) in the underlying container which behaves like a hashtable.
@@ -97,13 +97,13 @@ private Options() {
    * @return a new instance of MapStage
    */
   @Endpoint(describeByClass = true)
-  public static MapStage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, Iterable<Operand<?>> values, List<DataType<?>> dtypes, Options... options) {
+  public static MapStage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, Iterable<Operand<?>> values, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MapStage", scope.makeOpName("MapStage"));
     opBuilder.addInput(key.asOutput());
     opBuilder.addInput(indices.asOutput());
     opBuilder.addInputList(Operands.asOutputs(values));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstage.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstage.java
index 849f6f3ef6a..ba02210d5c4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstage.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstage.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -99,12 +98,12 @@ private Options() {
    * @return a new instance of MapUnstage
    */
   @Endpoint(describeByClass = true)
-  public static MapUnstage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<DataType<?>> dtypes, Options... options) {
+  public static MapUnstage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MapUnstage", scope.makeOpName("MapUnstage"));
     opBuilder.addInput(key.asOutput());
     opBuilder.addInput(indices.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstageNoKey.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstageNoKey.java
index 10a119ec6c4..f6cfe6c1d73 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstageNoKey.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MapUnstageNoKey.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -30,6 +29,7 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op removes and returns a random (key, value)
@@ -96,11 +96,11 @@ private Options() {
    * @return a new instance of MapUnstageNoKey
    */
   @Endpoint(describeByClass = true)
-  public static MapUnstageNoKey create(Scope scope, Operand<TInt32> indices, List<DataType<?>> dtypes, Options... options) {
+  public static MapUnstageNoKey create(Scope scope, Operand<TInt32> indices, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MapUnstageNoKey", scope.makeOpName("MapUnstageNoKey"));
     opBuilder.addInput(indices.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MlirPassthroughOp.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MlirPassthroughOp.java
index cc278fdae8f..027c75bf163 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MlirPassthroughOp.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MlirPassthroughOp.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -78,12 +77,12 @@ public final class MlirPassthroughOp extends RawOp implements Iterable<Operand<T
    * @return a new instance of MlirPassthroughOp
    */
   @Endpoint(describeByClass = true)
-  public static MlirPassthroughOp create(Scope scope, Iterable<Operand<?>> inputs, String mlirModule, List<DataType<?>> Toutputs) {
+  public static MlirPassthroughOp create(Scope scope, Iterable<Operand<?>> inputs, String mlirModule, List<Class<? extends TType>> Toutputs) {
     OperationBuilder opBuilder = scope.env().opBuilder("MlirPassthroughOp", scope.makeOpName("MlirPassthroughOp"));
     opBuilder.addInputList(Operands.asOutputs(inputs));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("mlir_module", mlirModule);
-    DataType[] ToutputsArray = new DataType[Toutputs.size()];
+    Class[] ToutputsArray = new Class[Toutputs.size()];
     for (int i = 0; i < ToutputsArray.length; ++i) {
       ToutputsArray[i] = Toutputs.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableDenseHashTable.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableDenseHashTable.java
index 4ff3c1f3ea6..0e6b35f502a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableDenseHashTable.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableDenseHashTable.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -122,7 +121,7 @@ private Options() {
    * @return a new instance of MutableDenseHashTable
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TType> MutableDenseHashTable create(Scope scope, Operand<T> emptyKey, Operand<T> deletedKey, DataType<U> valueDtype, Options... options) {
+  public static <T extends TType, U extends TType> MutableDenseHashTable create(Scope scope, Operand<T> emptyKey, Operand<T> deletedKey, Class<U> valueDtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MutableDenseHashTableV2", scope.makeOpName("MutableDenseHashTable"));
     opBuilder.addInput(emptyKey.asOutput());
     opBuilder.addInput(deletedKey.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTable.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTable.java
index a11789f9f34..0e2dd872331 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTable.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTable.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -88,7 +87,7 @@ private Options() {
    * @return a new instance of MutableHashTable
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TType> MutableHashTable create(Scope scope, DataType<T> keyDtype, DataType<U> valueDtype, Options... options) {
+  public static <T extends TType, U extends TType> MutableHashTable create(Scope scope, Class<T> keyDtype, Class<U> valueDtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MutableHashTableV2", scope.makeOpName("MutableHashTable"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("key_dtype", keyDtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTableOfTensors.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTableOfTensors.java
index 975f040ae3b..49d2e3c0373 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTableOfTensors.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/MutableHashTableOfTensors.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -97,7 +96,7 @@ private Options() {
    * @return a new instance of MutableHashTableOfTensors
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TType> MutableHashTableOfTensors create(Scope scope, DataType<T> keyDtype, DataType<U> valueDtype, Options... options) {
+  public static <T extends TType, U extends TType> MutableHashTableOfTensors create(Scope scope, Class<T> keyDtype, Class<U> valueDtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MutableHashTableOfTensorsV2", scope.makeOpName("MutableHashTableOfTensors"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("key_dtype", keyDtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclAllReduce.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclAllReduce.java
index a44611b5a41..171cf01d750 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclAllReduce.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclAllReduce.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs a tensor containing the reduction across all input tensors.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclBroadcast.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclBroadcast.java
index da0832e437a..17db313a42e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclBroadcast.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclBroadcast.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Sends `input` to all devices that are connected to the output.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclReduce.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclReduce.java
index ced1473e60a..6156dbc5551 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclReduce.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/NcclReduce.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Reduces `input` from `num_devices` using `reduction` to a single device.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapClear.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapClear.java
index 05a1b7ab984..df4b4f26bfc 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapClear.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapClear.java
@@ -18,13 +18,13 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
 import org.tensorflow.op.RawOp;
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op removes all elements in the underlying container.
@@ -87,10 +87,10 @@ private Options() {
    * @return a new instance of OrderedMapClear
    */
   @Endpoint(describeByClass = true)
-  public static OrderedMapClear create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static OrderedMapClear create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OrderedMapClear", scope.makeOpName("OrderedMapClear"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapIncompleteSize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapIncompleteSize.java
index 865810568db..1b686b2f14b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapIncompleteSize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapIncompleteSize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,6 +27,7 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op returns the number of incomplete elements in the underlying container.
@@ -90,10 +90,10 @@ private Options() {
    * @return a new instance of OrderedMapIncompleteSize
    */
   @Endpoint(describeByClass = true)
-  public static OrderedMapIncompleteSize create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static OrderedMapIncompleteSize create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OrderedMapIncompleteSize", scope.makeOpName("OrderedMapIncompleteSize"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapPeek.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapPeek.java
index 21c6adcc039..6b052e5f00c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapPeek.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapPeek.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -100,12 +99,12 @@ private Options() {
    * @return a new instance of OrderedMapPeek
    */
   @Endpoint(describeByClass = true)
-  public static OrderedMapPeek create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<DataType<?>> dtypes, Options... options) {
+  public static OrderedMapPeek create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OrderedMapPeek", scope.makeOpName("OrderedMapPeek"));
     opBuilder.addInput(key.asOutput());
     opBuilder.addInput(indices.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapSize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapSize.java
index afdee7de1bd..69c7851b13c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapSize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapSize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,6 +27,7 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op returns the number of elements in the underlying container.
@@ -90,10 +90,10 @@ private Options() {
    * @return a new instance of OrderedMapSize
    */
   @Endpoint(describeByClass = true)
-  public static OrderedMapSize create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static OrderedMapSize create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OrderedMapSize", scope.makeOpName("OrderedMapSize"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapStage.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapStage.java
index 7e02973e3c6..12f3155d661 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapStage.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapStage.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,6 +28,7 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TType;
 
 /**
  * Stage (key, values) in the underlying container which behaves like a ordered
@@ -99,13 +99,13 @@ private Options() {
    * @return a new instance of OrderedMapStage
    */
   @Endpoint(describeByClass = true)
-  public static OrderedMapStage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, Iterable<Operand<?>> values, List<DataType<?>> dtypes, Options... options) {
+  public static OrderedMapStage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, Iterable<Operand<?>> values, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OrderedMapStage", scope.makeOpName("OrderedMapStage"));
     opBuilder.addInput(key.asOutput());
     opBuilder.addInput(indices.asOutput());
     opBuilder.addInputList(Operands.asOutputs(values));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstage.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstage.java
index e2460e42dd8..8b08d7449c2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstage.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstage.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -99,12 +98,12 @@ private Options() {
    * @return a new instance of OrderedMapUnstage
    */
   @Endpoint(describeByClass = true)
-  public static OrderedMapUnstage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<DataType<?>> dtypes, Options... options) {
+  public static OrderedMapUnstage create(Scope scope, Operand<TInt64> key, Operand<TInt32> indices, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OrderedMapUnstage", scope.makeOpName("OrderedMapUnstage"));
     opBuilder.addInput(key.asOutput());
     opBuilder.addInput(indices.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstageNoKey.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstageNoKey.java
index f20a23b9806..edb9e82401a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstageNoKey.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/OrderedMapUnstageNoKey.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -30,6 +29,7 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op removes and returns the (key, value) element with the smallest
@@ -96,11 +96,11 @@ private Options() {
    * @return a new instance of OrderedMapUnstageNoKey
    */
   @Endpoint(describeByClass = true)
-  public static OrderedMapUnstageNoKey create(Scope scope, Operand<TInt32> indices, List<DataType<?>> dtypes, Options... options) {
+  public static OrderedMapUnstageNoKey create(Scope scope, Operand<TInt32> indices, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OrderedMapUnstageNoKey", scope.makeOpName("OrderedMapUnstageNoKey"));
     opBuilder.addInput(indices.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Placeholder.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Placeholder.java
index caef9fc0783..75d2326d007 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Placeholder.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Placeholder.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -70,7 +69,7 @@ private Options() {
    * @return a new instance of Placeholder
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> Placeholder<T> create(Scope scope, DataType<T> dtype, Options... options) {
+  public static <T extends TType> Placeholder<T> create(Scope scope, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Placeholder", scope.makeOpName("Placeholder"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Range.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Range.java
index 967a490382e..4f71eafe3c0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Range.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Range.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Creates a sequence of numbers.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReadVariableOp.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReadVariableOp.java
index aa2d1d2b9b2..6f33217ec56 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReadVariableOp.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReadVariableOp.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -52,7 +51,7 @@ public final class ReadVariableOp<T extends TType> extends RawOp implements Oper
    * @return a new instance of ReadVariableOp
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> ReadVariableOp<T> create(Scope scope, Operand<?> resource, DataType<T> dtype) {
+  public static <T extends TType> ReadVariableOp<T> create(Scope scope, Operand<?> resource, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("ReadVariableOp", scope.makeOpName("ReadVariableOp"));
     opBuilder.addInput(resource.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Recv.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Recv.java
index db6511c1583..81ebf79a0e8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Recv.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Recv.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -70,7 +69,7 @@ private Options() {
    * @return a new instance of Recv
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> Recv<T> create(Scope scope, DataType<T> tensorType, String tensorName, String sendDevice, Long sendDeviceIncarnation, String recvDevice, Options... options) {
+  public static <T extends TType> Recv<T> create(Scope scope, Class<T> tensorType, String tensorName, String sendDevice, Long sendDeviceIncarnation, String recvDevice, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Recv", scope.makeOpName("Recv"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("tensor_type", tensorType);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAll.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAll.java
index 9a5ad026ac8..1e940c95fb5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAll.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAll.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the "logical and" of elements across dimensions of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAny.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAny.java
index de479629f97..412e00fd07e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAny.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ReduceAny.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the "logical or" of elements across dimensions of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/RemoteFusedGraphExecute.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/RemoteFusedGraphExecute.java
index bd76549c976..f1a70aa35e6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/RemoteFusedGraphExecute.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/RemoteFusedGraphExecute.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -57,11 +56,11 @@ public final class RemoteFusedGraphExecute extends RawOp implements Iterable<Ope
    * @return a new instance of RemoteFusedGraphExecute
    */
   @Endpoint(describeByClass = true)
-  public static RemoteFusedGraphExecute create(Scope scope, Iterable<Operand<?>> inputs, List<DataType<?>> Toutputs, String serializedRemoteFusedGraphExecuteInfo) {
+  public static RemoteFusedGraphExecute create(Scope scope, Iterable<Operand<?>> inputs, List<Class<? extends TType>> Toutputs, String serializedRemoteFusedGraphExecuteInfo) {
     OperationBuilder opBuilder = scope.env().opBuilder("RemoteFusedGraphExecute", scope.makeOpName("RemoteFusedGraphExecute"));
     opBuilder.addInputList(Operands.asOutputs(inputs));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] ToutputsArray = new DataType[Toutputs.size()];
+    Class[] ToutputsArray = new Class[Toutputs.size()];
     for (int i = 0; i < ToutputsArray.length; ++i) {
       ToutputsArray[i] = Toutputs.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceCountUpTo.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceCountUpTo.java
index 19f630dd014..072c70a1862 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceCountUpTo.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceCountUpTo.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -27,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Increments variable pointed to by 'resource' until it reaches 'limit'.
@@ -48,7 +46,7 @@ public final class ResourceCountUpTo<T extends TNumber> extends RawOp implements
    * @return a new instance of ResourceCountUpTo
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> ResourceCountUpTo<T> create(Scope scope, Operand<?> resource, Long limit, DataType<T> T) {
+  public static <T extends TNumber> ResourceCountUpTo<T> create(Scope scope, Operand<?> resource, Long limit, Class<T> T) {
     OperationBuilder opBuilder = scope.env().opBuilder("ResourceCountUpTo", scope.makeOpName("ResourceCountUpTo"));
     opBuilder.addInput(resource.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGather.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGather.java
index 38aa1fbd407..d470cf09127 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGather.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGather.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -90,7 +89,7 @@ private Options() {
    * @return a new instance of ResourceGather
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TNumber> ResourceGather<U> create(Scope scope, Operand<?> resource, Operand<T> indices, DataType<U> dtype, Options... options) {
+  public static <U extends TType, T extends TNumber> ResourceGather<U> create(Scope scope, Operand<?> resource, Operand<T> indices, Class<U> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ResourceGather", scope.makeOpName("ResourceGather"));
     opBuilder.addInput(resource.asOutput());
     opBuilder.addInput(indices.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGatherNd.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGatherNd.java
index 85e422179f7..3f8237d98e6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGatherNd.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ResourceGatherNd.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,7 +44,7 @@ public final class ResourceGatherNd<U extends TType> extends RawOp implements Op
    * @return a new instance of ResourceGatherNd
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TNumber> ResourceGatherNd<U> create(Scope scope, Operand<?> resource, Operand<T> indices, DataType<U> dtype) {
+  public static <U extends TType, T extends TNumber> ResourceGatherNd<U> create(Scope scope, Operand<?> resource, Operand<T> indices, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("ResourceGatherNd", scope.makeOpName("ResourceGatherNd"));
     opBuilder.addInput(resource.asOutput());
     opBuilder.addInput(indices.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMax.java
index 20b84f016ea..360cb496ee8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMax.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Reduces sparse updates into a variable reference using the `max` operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMin.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMin.java
index 5010da14eaa..14f83e0756a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMin.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ScatterMin.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Reduces sparse updates into a variable reference using the `min` operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/SetDiff1d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/SetDiff1d.java
index c7e458931e5..d1c53319c4f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/SetDiff1d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/SetDiff1d.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -69,7 +68,7 @@ public final class SetDiff1d<T extends TType, U extends TNumber> extends RawOp {
    * @return a new instance of SetDiff1d
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TNumber> SetDiff1d<T, U> create(Scope scope, Operand<T> x, Operand<T> y, DataType<U> outIdx) {
+  public static <T extends TType, U extends TNumber> SetDiff1d<T, U> create(Scope scope, Operand<T> x, Operand<T> y, Class<U> outIdx) {
     OperationBuilder opBuilder = scope.env().opBuilder("ListDiff", scope.makeOpName("SetDiff1d"));
     opBuilder.addInput(x.asOutput());
     opBuilder.addInput(y.asOutput());
@@ -88,7 +87,7 @@ public static <T extends TType, U extends TNumber> SetDiff1d<T, U> create(Scope
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> SetDiff1d<T, TInt32> create(Scope scope, Operand<T> x, Operand<T> y) {
-    return create(scope, x, y, TInt32.DTYPE);
+    return create(scope, x, y, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Shape.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Shape.java
index 5d613401fc5..3ee92514a35 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Shape.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Shape.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -56,7 +55,7 @@ public final class Shape<U extends TNumber> extends RawOp implements Operand<U>
    * @return a new instance of Shape
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Shape<U> create(Scope scope, Operand<T> input, DataType<U> outType) {
+  public static <U extends TNumber, T extends TType> Shape<U> create(Scope scope, Operand<T> input, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("Shape", scope.makeOpName("Shape"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -73,7 +72,7 @@ public static <U extends TNumber, T extends TType> Shape<U> create(Scope scope,
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Shape<TInt32> create(Scope scope, Operand<T> input) {
-    return create(scope, input, TInt32.DTYPE);
+    return create(scope, input, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ShapeN.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ShapeN.java
index 42340a8a83d..1b1af62e7ac 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ShapeN.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/ShapeN.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -53,7 +52,7 @@ public final class ShapeN<U extends TNumber> extends RawOp implements Iterable<O
    * @return a new instance of ShapeN
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> ShapeN<U> create(Scope scope, Iterable<Operand<T>> input, DataType<U> outType) {
+  public static <U extends TNumber, T extends TType> ShapeN<U> create(Scope scope, Iterable<Operand<T>> input, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("ShapeN", scope.makeOpName("ShapeN"));
     opBuilder.addInputList(Operands.asOutputs(input));
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -70,7 +69,7 @@ public static <U extends TNumber, T extends TType> ShapeN<U> create(Scope scope,
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> ShapeN<TInt32> create(Scope scope, Iterable<Operand<T>> input) {
-    return create(scope, input, TInt32.DTYPE);
+    return create(scope, input, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Size.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Size.java
index 4aaccae0f30..ddd912f68f8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Size.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Size.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -57,7 +56,7 @@ public final class Size<U extends TNumber> extends RawOp implements Operand<U> {
    * @return a new instance of Size
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Size<U> create(Scope scope, Operand<T> input, DataType<U> outType) {
+  public static <U extends TNumber, T extends TType> Size<U> create(Scope scope, Operand<T> input, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("Size", scope.makeOpName("Size"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -74,7 +73,7 @@ public static <U extends TNumber, T extends TType> Size<U> create(Scope scope, O
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Size<TInt32> create(Scope scope, Operand<T> input) {
-    return create(scope, input, TInt32.DTYPE);
+    return create(scope, input, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageClear.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageClear.java
index 60e51559f74..1bb2fde0fdf 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageClear.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageClear.java
@@ -18,13 +18,13 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
 import org.tensorflow.op.RawOp;
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op removes all elements in the underlying container.
@@ -87,10 +87,10 @@ private Options() {
    * @return a new instance of StageClear
    */
   @Endpoint(describeByClass = true)
-  public static StageClear create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static StageClear create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("StageClear", scope.makeOpName("StageClear"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StagePeek.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StagePeek.java
index 2126de722df..07ce640c947 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StagePeek.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StagePeek.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -98,11 +97,11 @@ private Options() {
    * @return a new instance of StagePeek
    */
   @Endpoint(describeByClass = true)
-  public static StagePeek create(Scope scope, Operand<TInt32> index, List<DataType<?>> dtypes, Options... options) {
+  public static StagePeek create(Scope scope, Operand<TInt32> index, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("StagePeek", scope.makeOpName("StagePeek"));
     opBuilder.addInput(index.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageSize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageSize.java
index 94ef566e708..f67f23e5bd3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageSize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/StageSize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,6 +27,7 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
+import org.tensorflow.types.family.TType;
 
 /**
  * Op returns the number of elements in the underlying container.
@@ -90,10 +90,10 @@ private Options() {
    * @return a new instance of StageSize
    */
   @Endpoint(describeByClass = true)
-  public static StageSize create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static StageSize create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("StageSize", scope.makeOpName("StageSize"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TemporaryVariable.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TemporaryVariable.java
index 43d9247fe21..452d380f859 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TemporaryVariable.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TemporaryVariable.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -82,7 +81,7 @@ private Options() {
    * @return a new instance of TemporaryVariable
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TemporaryVariable<T> create(Scope scope, Shape shape, DataType<T> dtype, Options... options) {
+  public static <T extends TType> TemporaryVariable<T> create(Scope scope, Shape shape, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TemporaryVariable", scope.makeOpName("TemporaryVariable"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("shape", shape);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArray.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArray.java
index f34dc414484..fb6014a5efd 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArray.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArray.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -116,7 +115,7 @@ private Options() {
    * @return a new instance of TensorArray
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorArray create(Scope scope, Operand<TInt32> size, DataType<T> dtype, Options... options) {
+  public static <T extends TType> TensorArray create(Scope scope, Operand<TInt32> size, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorArrayV3", scope.makeOpName("TensorArray"));
     opBuilder.addInput(size.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayConcat.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayConcat.java
index e25b7a2f958..47d78d77f1d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayConcat.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayConcat.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -83,7 +82,7 @@ private Options() {
    * @return a new instance of TensorArrayConcat
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorArrayConcat<T> create(Scope scope, Operand<?> handle, Operand<TFloat32> flowIn, DataType<T> dtype, Options... options) {
+  public static <T extends TType> TensorArrayConcat<T> create(Scope scope, Operand<?> handle, Operand<TFloat32> flowIn, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorArrayConcatV3", scope.makeOpName("TensorArrayConcat"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(flowIn.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayGather.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayGather.java
index 13c6dce3122..9ba3bca49f7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayGather.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayGather.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -74,7 +73,7 @@ private Options() {
    * @return a new instance of TensorArrayGather
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorArrayGather<T> create(Scope scope, Operand<?> handle, Operand<TInt32> indices, Operand<TFloat32> flowIn, DataType<T> dtype, Options... options) {
+  public static <T extends TType> TensorArrayGather<T> create(Scope scope, Operand<?> handle, Operand<TInt32> indices, Operand<TFloat32> flowIn, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorArrayGatherV3", scope.makeOpName("TensorArrayGather"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(indices.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayPack.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayPack.java
index 4b4be4ac089..d622be4d3e7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayPack.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayPack.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -67,7 +66,7 @@ private Options() {
    * @return a new instance of TensorArrayPack
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorArrayPack<T> create(Scope scope, Operand<TString> handle, Operand<TFloat32> flowIn, DataType<T> dtype, Options... options) {
+  public static <T extends TType> TensorArrayPack<T> create(Scope scope, Operand<TString> handle, Operand<TFloat32> flowIn, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorArrayPack", scope.makeOpName("TensorArrayPack"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(flowIn.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayRead.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayRead.java
index 55cf48bb020..b71aa6c58a8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayRead.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorArrayRead.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,7 +48,7 @@ public final class TensorArrayRead<T extends TType> extends RawOp implements Ope
    * @return a new instance of TensorArrayRead
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorArrayRead<T> create(Scope scope, Operand<?> handle, Operand<TInt32> index, Operand<TFloat32> flowIn, DataType<T> dtype) {
+  public static <T extends TType> TensorArrayRead<T> create(Scope scope, Operand<?> handle, Operand<TInt32> index, Operand<TFloat32> flowIn, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorArrayReadV3", scope.makeOpName("TensorArrayRead"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(index.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcat.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcat.java
index 2f1771797e0..6f83df9232c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcat.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcat.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -62,7 +61,7 @@ public final class TensorListConcat<U extends TType> extends RawOp {
    * @return a new instance of TensorListConcat
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TNumber> TensorListConcat<U> create(Scope scope, Operand<?> inputHandle, Operand<T> elementShape, Operand<TInt64> leadingDims, DataType<U> elementDtype) {
+  public static <U extends TType, T extends TNumber> TensorListConcat<U> create(Scope scope, Operand<?> inputHandle, Operand<T> elementShape, Operand<TInt64> leadingDims, Class<U> elementDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListConcatV2", scope.makeOpName("TensorListConcat"));
     opBuilder.addInput(inputHandle.asOutput());
     opBuilder.addInput(elementShape.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcatLists.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcatLists.java
index fdca8e2d6cd..285bc6e734a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcatLists.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListConcatLists.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -43,7 +42,7 @@ public final class TensorListConcatLists extends RawOp implements Operand<TType>
    * @return a new instance of TensorListConcatLists
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorListConcatLists create(Scope scope, Operand<?> inputA, Operand<?> inputB, DataType<T> elementDtype) {
+  public static <T extends TType> TensorListConcatLists create(Scope scope, Operand<?> inputA, Operand<?> inputB, Class<T> elementDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListConcatLists", scope.makeOpName("TensorListConcatLists"));
     opBuilder.addInput(inputA.asOutput());
     opBuilder.addInput(inputB.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListElementShape.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListElementShape.java
index 2c3e3a5b90c..17dfddcb958 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListElementShape.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListElementShape.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -27,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * The shape of the elements of the given list, as a tensor.
@@ -49,7 +47,7 @@ public final class TensorListElementShape<T extends TNumber> extends RawOp imple
    * @return a new instance of TensorListElementShape
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> TensorListElementShape<T> create(Scope scope, Operand<?> inputHandle, DataType<T> shapeType) {
+  public static <T extends TNumber> TensorListElementShape<T> create(Scope scope, Operand<?> inputHandle, Class<T> shapeType) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListElementShape", scope.makeOpName("TensorListElementShape"));
     opBuilder.addInput(inputHandle.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGather.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGather.java
index dbb565a29a6..d99e658f936 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGather.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGather.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -55,7 +54,7 @@ public final class TensorListGather<T extends TType> extends RawOp implements Op
    * @return a new instance of TensorListGather
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorListGather<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> indices, Operand<TInt32> elementShape, DataType<T> elementDtype) {
+  public static <T extends TType> TensorListGather<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> indices, Operand<TInt32> elementShape, Class<T> elementDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListGather", scope.makeOpName("TensorListGather"));
     opBuilder.addInput(inputHandle.asOutput());
     opBuilder.addInput(indices.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGetItem.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGetItem.java
index 9af925e1425..ed92723ab9f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGetItem.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListGetItem.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,7 +45,7 @@ public final class TensorListGetItem<T extends TType> extends RawOp implements O
    * @return a new instance of TensorListGetItem
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorListGetItem<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> index, Operand<TInt32> elementShape, DataType<T> elementDtype) {
+  public static <T extends TType> TensorListGetItem<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> index, Operand<TInt32> elementShape, Class<T> elementDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListGetItem", scope.makeOpName("TensorListGetItem"));
     opBuilder.addInput(inputHandle.asOutput());
     opBuilder.addInput(index.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListPopBack.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListPopBack.java
index 96fd3433e07..1cb362c913c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListPopBack.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListPopBack.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -54,7 +53,7 @@ public final class TensorListPopBack<T extends TType> extends RawOp {
    * @return a new instance of TensorListPopBack
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorListPopBack<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> elementShape, DataType<T> elementDtype) {
+  public static <T extends TType> TensorListPopBack<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> elementShape, Class<T> elementDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListPopBack", scope.makeOpName("TensorListPopBack"));
     opBuilder.addInput(inputHandle.asOutput());
     opBuilder.addInput(elementShape.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListReserve.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListReserve.java
index 3f798e0f998..e7c6abda2c5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListReserve.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListReserve.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,7 +50,7 @@ public final class TensorListReserve extends RawOp implements Operand<TType> {
    * @return a new instance of TensorListReserve
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber, U extends TType> TensorListReserve create(Scope scope, Operand<T> elementShape, Operand<TInt32> numElements, DataType<U> elementDtype) {
+  public static <T extends TNumber, U extends TType> TensorListReserve create(Scope scope, Operand<T> elementShape, Operand<TInt32> numElements, Class<U> elementDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListReserve", scope.makeOpName("TensorListReserve"));
     opBuilder.addInput(elementShape.asOutput());
     opBuilder.addInput(numElements.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListStack.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListStack.java
index c0ecf388b3d..dfc2357d18c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListStack.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorListStack.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -74,7 +73,7 @@ private Options() {
    * @return a new instance of TensorListStack
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TensorListStack<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> elementShape, DataType<T> elementDtype, Options... options) {
+  public static <T extends TType> TensorListStack<T> create(Scope scope, Operand<?> inputHandle, Operand<TInt32> elementShape, Class<T> elementDtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TensorListStack", scope.makeOpName("TensorListStack"));
     opBuilder.addInput(inputHandle.asOutput());
     opBuilder.addInput(elementShape.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorScatterMax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorScatterMax.java
deleted file mode 100644
index 1a3042926f6..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorScatterMax.java
+++ /dev/null
@@ -1,75 +0,0 @@
-/* Copyright 2018 The TensorFlow Authors. All Rights Reserved.
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-=======================================================================*/
-
-// This class has been generated, DO NOT EDIT!
-
-package org.tensorflow.op.core;
-
-import org.tensorflow.Operand;
-import org.tensorflow.Operation;
-import org.tensorflow.OperationBuilder;
-import org.tensorflow.Output;
-import org.tensorflow.op.RawOp;
-import org.tensorflow.op.Scope;
-import org.tensorflow.op.annotation.Endpoint;
-import org.tensorflow.op.annotation.Operator;
-import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
-
-/**
- * @param <T> data type for {@code output()} output
- */
-@Operator
-public final class TensorScatterMax<T extends TType> extends RawOp implements Operand<T> {
-  
-  /**
-   * Factory method to create a class wrapping a new TensorScatterMax operation.
-   * 
-   * @param scope current scope
-   * @param tensor Tensor to update.
-   * @param indices Index tensor.
-   * @param updates Updates to scatter into output.
-   * @return a new instance of TensorScatterMax
-   */
-  @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TNumber> TensorScatterMax<T> create(Scope scope, Operand<T> tensor, Operand<U> indices, Operand<T> updates) {
-    OperationBuilder opBuilder = scope.env().opBuilder("TensorScatterMax", scope.makeOpName("TensorScatterMax"));
-    opBuilder.addInput(tensor.asOutput());
-    opBuilder.addInput(indices.asOutput());
-    opBuilder.addInput(updates.asOutput());
-    opBuilder = scope.applyControlDependencies(opBuilder);
-    return new TensorScatterMax<T>(opBuilder.build());
-  }
-  
-  /**
-   * A new tensor copied from tensor whose values are element-wise maximum between tensor and updates according to the indices.
-   */
-  public Output<T> output() {
-    return output;
-  }
-  
-  @Override
-  public Output<T> asOutput() {
-    return output;
-  }
-  
-  private Output<T> output;
-  
-  private TensorScatterMax(Operation operation) {
-    super(operation);
-    int outputIdx = 0;
-    output = operation.output(outputIdx++);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorScatterMin.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorScatterMin.java
deleted file mode 100644
index 5f7b0d0eb77..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/TensorScatterMin.java
+++ /dev/null
@@ -1,75 +0,0 @@
-/* Copyright 2018 The TensorFlow Authors. All Rights Reserved.
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-=======================================================================*/
-
-// This class has been generated, DO NOT EDIT!
-
-package org.tensorflow.op.core;
-
-import org.tensorflow.Operand;
-import org.tensorflow.Operation;
-import org.tensorflow.OperationBuilder;
-import org.tensorflow.Output;
-import org.tensorflow.op.RawOp;
-import org.tensorflow.op.Scope;
-import org.tensorflow.op.annotation.Endpoint;
-import org.tensorflow.op.annotation.Operator;
-import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
-
-/**
- * @param <T> data type for {@code output()} output
- */
-@Operator
-public final class TensorScatterMin<T extends TType> extends RawOp implements Operand<T> {
-  
-  /**
-   * Factory method to create a class wrapping a new TensorScatterMin operation.
-   * 
-   * @param scope current scope
-   * @param tensor Tensor to update.
-   * @param indices Index tensor.
-   * @param updates Updates to scatter into output.
-   * @return a new instance of TensorScatterMin
-   */
-  @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TNumber> TensorScatterMin<T> create(Scope scope, Operand<T> tensor, Operand<U> indices, Operand<T> updates) {
-    OperationBuilder opBuilder = scope.env().opBuilder("TensorScatterMin", scope.makeOpName("TensorScatterMin"));
-    opBuilder.addInput(tensor.asOutput());
-    opBuilder.addInput(indices.asOutput());
-    opBuilder.addInput(updates.asOutput());
-    opBuilder = scope.applyControlDependencies(opBuilder);
-    return new TensorScatterMin<T>(opBuilder.build());
-  }
-  
-  /**
-   * A new tensor copied from tensor whose values are element-wise minimum between tensor and updates according to the indices.
-   */
-  public Output<T> output() {
-    return output;
-  }
-  
-  @Override
-  public Output<T> asOutput() {
-    return output;
-  }
-  
-  private Output<T> output;
-  
-  private TensorScatterMin(Operation operation) {
-    super(operation);
-    int outputIdx = 0;
-    output = operation.output(outputIdx++);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unique.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unique.java
index f3b8b04aead..8d1bcc7a6ef 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unique.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unique.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -90,7 +89,7 @@ public final class Unique<T extends TType, V extends TNumber> extends RawOp {
    * @return a new instance of Unique
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, V extends TNumber, U extends TNumber> Unique<T, V> create(Scope scope, Operand<T> x, Operand<U> axis, DataType<V> outIdx) {
+  public static <T extends TType, V extends TNumber, U extends TNumber> Unique<T, V> create(Scope scope, Operand<T> x, Operand<U> axis, Class<V> outIdx) {
     OperationBuilder opBuilder = scope.env().opBuilder("UniqueV2", scope.makeOpName("Unique"));
     opBuilder.addInput(x.asOutput());
     opBuilder.addInput(axis.asOutput());
@@ -110,7 +109,7 @@ public static <T extends TType, V extends TNumber, U extends TNumber> Unique<T,
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType, U extends TNumber> Unique<T, TInt32> create(Scope scope, Operand<T> x, Operand<U> axis) {
-    return create(scope, x, axis, TInt32.DTYPE);
+    return create(scope, x, axis, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UniqueWithCounts.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UniqueWithCounts.java
index 732d4432333..6316f3659c9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UniqueWithCounts.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UniqueWithCounts.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -94,7 +93,7 @@ public final class UniqueWithCounts<T extends TType, V extends TNumber> extends
    * @return a new instance of UniqueWithCounts
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, V extends TNumber, U extends TNumber> UniqueWithCounts<T, V> create(Scope scope, Operand<T> x, Operand<U> axis, DataType<V> outIdx) {
+  public static <T extends TType, V extends TNumber, U extends TNumber> UniqueWithCounts<T, V> create(Scope scope, Operand<T> x, Operand<U> axis, Class<V> outIdx) {
     OperationBuilder opBuilder = scope.env().opBuilder("UniqueWithCountsV2", scope.makeOpName("UniqueWithCounts"));
     opBuilder.addInput(x.asOutput());
     opBuilder.addInput(axis.asOutput());
@@ -114,7 +113,7 @@ public static <T extends TType, V extends TNumber, U extends TNumber> UniqueWith
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType, U extends TNumber> UniqueWithCounts<T, TInt32> create(Scope scope, Operand<T> x, Operand<U> axis) {
-    return create(scope, x, axis, TInt32.DTYPE);
+    return create(scope, x, axis, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UnravelIndex.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UnravelIndex.java
index ecc073e37a4..37c5f2ded90 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UnravelIndex.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UnravelIndex.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Converts an array of flat indices into a tuple of coordinate arrays.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unstage.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unstage.java
index 1878ce6fb88..f6bd852f898 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unstage.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Unstage.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -95,10 +94,10 @@ private Options() {
    * @return a new instance of Unstage
    */
   @Endpoint(describeByClass = true)
-  public static Unstage create(Scope scope, List<DataType<?>> dtypes, Options... options) {
+  public static Unstage create(Scope scope, List<Class<? extends TType>> dtypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Unstage", scope.makeOpName("Unstage"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UpperBound.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UpperBound.java
index c46fbcba4de..67d2e9255e4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UpperBound.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/UpperBound.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -66,7 +65,7 @@ public final class UpperBound<U extends TNumber> extends RawOp implements Operan
    * @return a new instance of UpperBound
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> UpperBound<U> create(Scope scope, Operand<T> sortedInputs, Operand<T> values, DataType<U> outType) {
+  public static <U extends TNumber, T extends TType> UpperBound<U> create(Scope scope, Operand<T> sortedInputs, Operand<T> values, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("UpperBound", scope.makeOpName("UpperBound"));
     opBuilder.addInput(sortedInputs.asOutput());
     opBuilder.addInput(values.asOutput());
@@ -86,7 +85,7 @@ public static <U extends TNumber, T extends TType> UpperBound<U> create(Scope sc
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> UpperBound<TInt32> create(Scope scope, Operand<T> sortedInputs, Operand<T> values) {
-    return create(scope, sortedInputs, values, TInt32.DTYPE);
+    return create(scope, sortedInputs, values, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VarHandleOp.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VarHandleOp.java
index 92618c33d37..c264519b454 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VarHandleOp.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VarHandleOp.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.core;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -85,7 +84,7 @@ private Options() {
    * @return a new instance of VarHandleOp
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> VarHandleOp create(Scope scope, DataType<T> dtype, Shape shape, Options... options) {
+  public static <T extends TType> VarHandleOp create(Scope scope, Class<T> dtype, Shape shape, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("VarHandleOp", scope.makeOpName("VarHandleOp"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Variable.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Variable.java
index 7353dfee688..bcd9719c932 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Variable.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/Variable.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -81,7 +80,7 @@ private Options() {
    * @return a new instance of Variable
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> Variable<T> create(Scope scope, Shape shape, DataType<T> dtype, Options... options) {
+  public static <T extends TType> Variable<T> create(Scope scope, Shape shape, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("VariableV2", scope.makeOpName("Variable"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("shape", shape);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VariableShape.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VariableShape.java
index 6573cb8de11..2a4c0ba60ec 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VariableShape.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/core/VariableShape.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the shape of the variable pointed to by `resource`.
@@ -56,7 +54,7 @@ public final class VariableShape<T extends TNumber> extends RawOp implements Ope
    * @return a new instance of VariableShape
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> VariableShape<T> create(Scope scope, Operand<?> input, DataType<T> outType) {
+  public static <T extends TNumber> VariableShape<T> create(Scope scope, Operand<?> input, Class<T> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("VariableShape", scope.makeOpName("VariableShape"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -73,7 +71,7 @@ public static <T extends TNumber> VariableShape<T> create(Scope scope, Operand<?
    */
   @Endpoint(describeByClass = true)
   public static VariableShape<TInt32> create(Scope scope, Operand<?> input) {
-    return create(scope, input, TInt32.DTYPE);
+    return create(scope, input, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousIterator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousIterator.java
index bd3d1a400f5..8236e750ff5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousIterator.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousIterator.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
 import org.tensorflow.Output;
@@ -27,6 +26,7 @@
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
+import org.tensorflow.types.family.TType;
 
 /**
  * A container for an iterator resource.
@@ -43,10 +43,10 @@ public final class AnonymousIterator extends RawOp {
    * @return a new instance of AnonymousIterator
    */
   @Endpoint(describeByClass = true)
-  public static AnonymousIterator create(Scope scope, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static AnonymousIterator create(Scope scope, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("AnonymousIteratorV2", scope.makeOpName("AnonymousIterator"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousMultiDeviceIterator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousMultiDeviceIterator.java
index bff57b33c8f..9b5ba5b7fd9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousMultiDeviceIterator.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AnonymousMultiDeviceIterator.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
 import org.tensorflow.Output;
@@ -27,6 +26,7 @@
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
+import org.tensorflow.types.family.TType;
 
 /**
  * A container for a multi device iterator resource.
@@ -43,7 +43,7 @@ public final class AnonymousMultiDeviceIterator extends RawOp {
    * @return a new instance of AnonymousMultiDeviceIterator
    */
   @Endpoint(describeByClass = true)
-  public static AnonymousMultiDeviceIterator create(Scope scope, List<String> devices, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static AnonymousMultiDeviceIterator create(Scope scope, List<String> devices, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("AnonymousMultiDeviceIterator", scope.makeOpName("AnonymousMultiDeviceIterator"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     String[] devicesArray = new String[devices.size()];
@@ -51,7 +51,7 @@ public static AnonymousMultiDeviceIterator create(Scope scope, List<String> devi
       devicesArray[i] = devices.get(i);
     }
     opBuilder.setAttr("devices", devicesArray);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AssertNextDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AssertNextDataset.java
index c2825e29b93..0fd73311cf3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AssertNextDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AssertNextDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -58,12 +57,12 @@ public final class AssertNextDataset extends RawOp implements Operand<TType> {
    * @return a new instance of AssertNextDataset
    */
   @Endpoint(describeByClass = true)
-  public static AssertNextDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> transformations, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static AssertNextDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> transformations, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("AssertNextDataset", scope.makeOpName("AssertNextDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(transformations.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AutoShardDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AutoShardDataset.java
index aca90031261..c71a1b96f5e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AutoShardDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/AutoShardDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -76,13 +75,13 @@ private Options() {
    * @return a new instance of AutoShardDataset
    */
   @Endpoint(describeByClass = true)
-  public static AutoShardDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numWorkers, Operand<TInt64> index, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static AutoShardDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numWorkers, Operand<TInt64> index, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("AutoShardDataset", scope.makeOpName("AutoShardDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numWorkers.asOutput());
     opBuilder.addInput(index.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BatchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BatchDataset.java
index b0fd6ef0c0c..7409bdade59 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BatchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BatchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -71,13 +70,13 @@ private Options() {
    * @return a new instance of BatchDataset
    */
   @Endpoint(describeByClass = true)
-  public static BatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> batchSize, Operand<TBool> dropRemainder, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static BatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> batchSize, Operand<TBool> dropRemainder, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("BatchDatasetV2", scope.makeOpName("BatchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(batchSize.asOutput());
     opBuilder.addInput(dropRemainder.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BytesProducedStatsDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BytesProducedStatsDataset.java
index a3b95d92909..4000a238b86 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BytesProducedStatsDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/BytesProducedStatsDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class BytesProducedStatsDataset extends RawOp implements Operand<TT
    * @return a new instance of BytesProducedStatsDataset
    */
   @Endpoint(describeByClass = true)
-  public static BytesProducedStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static BytesProducedStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("BytesProducedStatsDataset", scope.makeOpName("BytesProducedStatsDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(tag.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDataset.java
index 07384fb6b3e..1de01e0eccd 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -53,12 +52,12 @@ public final class CacheDataset extends RawOp implements Operand<TType> {
    * @return a new instance of CacheDataset
    */
   @Endpoint(describeByClass = true)
-  public static CacheDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> filename, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static CacheDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> filename, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("CacheDataset", scope.makeOpName("CacheDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(filename.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDatasetV2.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDatasetV2.java
index ba370108e04..3bcab52b253 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDatasetV2.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/CacheDatasetV2.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,13 +46,13 @@ public final class CacheDatasetV2 extends RawOp implements Operand<TType> {
    * @return a new instance of CacheDatasetV2
    */
   @Endpoint(describeByClass = true)
-  public static CacheDatasetV2 create(Scope scope, Operand<?> inputDataset, Operand<TString> filename, Operand<?> cache, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static CacheDatasetV2 create(Scope scope, Operand<?> inputDataset, Operand<TString> filename, Operand<?> cache, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("CacheDatasetV2", scope.makeOpName("CacheDatasetV2"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(filename.asOutput());
     opBuilder.addInput(cache.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ChooseFastestDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ChooseFastestDataset.java
index a8d9ce445fe..74f34bb8023 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ChooseFastestDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ChooseFastestDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class ChooseFastestDataset extends RawOp implements Operand<TType>
    * @return a new instance of ChooseFastestDataset
    */
   @Endpoint(describeByClass = true)
-  public static ChooseFastestDataset create(Scope scope, Iterable<Operand<?>> inputDatasets, Long numExperiments, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static ChooseFastestDataset create(Scope scope, Iterable<Operand<?>> inputDatasets, Long numExperiments, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ChooseFastestDataset", scope.makeOpName("ChooseFastestDataset"));
     opBuilder.addInputList(Operands.asOutputs(inputDatasets));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("num_experiments", numExperiments);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ConcatenateDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ConcatenateDataset.java
index 19cf0b4a706..34326835089 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ConcatenateDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ConcatenateDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class ConcatenateDataset extends RawOp implements Operand<TType> {
    * @return a new instance of ConcatenateDataset
    */
   @Endpoint(describeByClass = true)
-  public static ConcatenateDataset create(Scope scope, Operand<?> inputDataset, Operand<?> anotherDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static ConcatenateDataset create(Scope scope, Operand<?> inputDataset, Operand<?> anotherDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ConcatenateDataset", scope.makeOpName("ConcatenateDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(anotherDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DatasetToSingleElement.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DatasetToSingleElement.java
index 91fd5032439..64a20123b56 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DatasetToSingleElement.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DatasetToSingleElement.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,11 +46,11 @@ public final class DatasetToSingleElement extends RawOp implements Iterable<Oper
    * @return a new instance of DatasetToSingleElement
    */
   @Endpoint(describeByClass = true)
-  public static DatasetToSingleElement create(Scope scope, Operand<?> dataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static DatasetToSingleElement create(Scope scope, Operand<?> dataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("DatasetToSingleElement", scope.makeOpName("DatasetToSingleElement"));
     opBuilder.addInput(dataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DenseToSparseBatchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DenseToSparseBatchDataset.java
index 32bd135c325..1882f3e8beb 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DenseToSparseBatchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DenseToSparseBatchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,13 +50,13 @@ public final class DenseToSparseBatchDataset extends RawOp implements Operand<TT
    * @return a new instance of DenseToSparseBatchDataset
    */
   @Endpoint(describeByClass = true)
-  public static DenseToSparseBatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> batchSize, Operand<TInt64> rowShape, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static DenseToSparseBatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> batchSize, Operand<TInt64> rowShape, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("DenseToSparseBatchDataset", scope.makeOpName("DenseToSparseBatchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(batchSize.asOutput());
     opBuilder.addInput(rowShape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DirectedInterleaveDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DirectedInterleaveDataset.java
index 6883e2218d9..4bf4ecdf141 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DirectedInterleaveDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/DirectedInterleaveDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,12 +48,12 @@ public final class DirectedInterleaveDataset extends RawOp implements Operand<TT
    * @return a new instance of DirectedInterleaveDataset
    */
   @Endpoint(describeByClass = true)
-  public static DirectedInterleaveDataset create(Scope scope, Operand<?> selectorInputDataset, Iterable<Operand<?>> dataInputDatasets, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static DirectedInterleaveDataset create(Scope scope, Operand<?> selectorInputDataset, Iterable<Operand<?>> dataInputDatasets, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("DirectedInterleaveDataset", scope.makeOpName("DirectedInterleaveDataset"));
     opBuilder.addInput(selectorInputDataset.asOutput());
     opBuilder.addInputList(Operands.asOutputs(dataInputDatasets));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/FilterByLastComponentDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/FilterByLastComponentDataset.java
index d6c2b576d7e..ea090e46fc6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/FilterByLastComponentDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/FilterByLastComponentDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class FilterByLastComponentDataset extends RawOp implements Operand
    * @return a new instance of FilterByLastComponentDataset
    */
   @Endpoint(describeByClass = true)
-  public static FilterByLastComponentDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static FilterByLastComponentDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("FilterByLastComponentDataset", scope.makeOpName("FilterByLastComponentDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IgnoreErrorsDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IgnoreErrorsDataset.java
index 7ffca90d150..08aaf2228a1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IgnoreErrorsDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IgnoreErrorsDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class IgnoreErrorsDataset extends RawOp implements Operand<TType> {
    * @return a new instance of IgnoreErrorsDataset
    */
   @Endpoint(describeByClass = true)
-  public static IgnoreErrorsDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static IgnoreErrorsDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("IgnoreErrorsDataset", scope.makeOpName("IgnoreErrorsDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/Iterator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/Iterator.java
index 45030001714..860c089fd40 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/Iterator.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/Iterator.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class Iterator extends RawOp implements Operand<TType> {
    * @return a new instance of Iterator
    */
   @Endpoint(describeByClass = true)
-  public static Iterator create(Scope scope, String sharedName, String container, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static Iterator create(Scope scope, String sharedName, String container, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("IteratorV2", scope.makeOpName("Iterator"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("shared_name", sharedName);
     opBuilder.setAttr("container", container);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorFromStringHandle.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorFromStringHandle.java
index 8264419635a..db50d961930 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorFromStringHandle.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorFromStringHandle.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -64,11 +63,11 @@ private Options() {
    * @return a new instance of IteratorFromStringHandle
    */
   @Endpoint(describeByClass = true)
-  public static IteratorFromStringHandle create(Scope scope, Operand<TString> stringHandle, List<DataType<?>> outputTypes, Options... options) {
+  public static IteratorFromStringHandle create(Scope scope, Operand<TString> stringHandle, List<Class<? extends TType>> outputTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("IteratorFromStringHandleV2", scope.makeOpName("IteratorFromStringHandle"));
     opBuilder.addInput(stringHandle.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNext.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNext.java
index b7be406e7bd..2a02a9dc1ea 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNext.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNext.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,11 +47,11 @@ public final class IteratorGetNext extends RawOp implements Iterable<Operand<TTy
    * @return a new instance of IteratorGetNext
    */
   @Endpoint(describeByClass = true)
-  public static IteratorGetNext create(Scope scope, Operand<?> iterator, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static IteratorGetNext create(Scope scope, Operand<?> iterator, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("IteratorGetNext", scope.makeOpName("IteratorGetNext"));
     opBuilder.addInput(iterator.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextAsOptional.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextAsOptional.java
index 682b7ba6f35..73d16c7b56f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextAsOptional.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextAsOptional.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,11 +45,11 @@ public final class IteratorGetNextAsOptional extends RawOp implements Operand<TT
    * @return a new instance of IteratorGetNextAsOptional
    */
   @Endpoint(describeByClass = true)
-  public static IteratorGetNextAsOptional create(Scope scope, Operand<?> iterator, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static IteratorGetNextAsOptional create(Scope scope, Operand<?> iterator, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("IteratorGetNextAsOptional", scope.makeOpName("IteratorGetNextAsOptional"));
     opBuilder.addInput(iterator.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextSync.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextSync.java
index 3cb4f072307..27a088fdf67 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextSync.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/IteratorGetNextSync.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -53,11 +52,11 @@ public final class IteratorGetNextSync extends RawOp implements Iterable<Operand
    * @return a new instance of IteratorGetNextSync
    */
   @Endpoint(describeByClass = true)
-  public static IteratorGetNextSync create(Scope scope, Operand<?> iterator, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static IteratorGetNextSync create(Scope scope, Operand<?> iterator, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("IteratorGetNextSync", scope.makeOpName("IteratorGetNextSync"));
     opBuilder.addInput(iterator.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LMDBDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LMDBDataset.java
index d0057d60323..042e866f469 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LMDBDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LMDBDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -58,11 +57,11 @@ public final class LMDBDataset extends RawOp implements Operand<TType> {
    * @return a new instance of LMDBDataset
    */
   @Endpoint(describeByClass = true)
-  public static LMDBDataset create(Scope scope, Operand<TString> filenames, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static LMDBDataset create(Scope scope, Operand<TString> filenames, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("LMDBDataset", scope.makeOpName("LMDBDataset"));
     opBuilder.addInput(filenames.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LatencyStatsDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LatencyStatsDataset.java
index 731b70867a7..29b9ba70579 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LatencyStatsDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LatencyStatsDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class LatencyStatsDataset extends RawOp implements Operand<TType> {
    * @return a new instance of LatencyStatsDataset
    */
   @Endpoint(describeByClass = true)
-  public static LatencyStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static LatencyStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("LatencyStatsDataset", scope.makeOpName("LatencyStatsDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(tag.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LeakyReluGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LeakyReluGrad.java
index 3fb21bd4a6e..e0c06a9b3b6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LeakyReluGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/LeakyReluGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes rectified linear gradients for a LeakyRelu operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MaxIntraOpParallelismDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MaxIntraOpParallelismDataset.java
index d7c67924b55..390c6d281ac 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MaxIntraOpParallelismDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MaxIntraOpParallelismDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class MaxIntraOpParallelismDataset extends RawOp implements Operand
    * @return a new instance of MaxIntraOpParallelismDataset
    */
   @Endpoint(describeByClass = true)
-  public static MaxIntraOpParallelismDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> maxIntraOpParallelism, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static MaxIntraOpParallelismDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> maxIntraOpParallelism, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("MaxIntraOpParallelismDataset", scope.makeOpName("MaxIntraOpParallelismDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(maxIntraOpParallelism.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ModelDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ModelDataset.java
index 53b6f10204d..fa1acc9e5b2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ModelDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ModelDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -76,11 +75,11 @@ private Options() {
    * @return a new instance of ModelDataset
    */
   @Endpoint(describeByClass = true)
-  public static ModelDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static ModelDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ModelDataset", scope.makeOpName("ModelDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIterator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIterator.java
index a3dfeeb5665..cb223414fec 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIterator.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIterator.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,7 +48,7 @@ public final class MultiDeviceIterator extends RawOp implements Operand<TType> {
    * @return a new instance of MultiDeviceIterator
    */
   @Endpoint(describeByClass = true)
-  public static MultiDeviceIterator create(Scope scope, List<String> devices, String sharedName, String container, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static MultiDeviceIterator create(Scope scope, List<String> devices, String sharedName, String container, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("MultiDeviceIterator", scope.makeOpName("MultiDeviceIterator"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     String[] devicesArray = new String[devices.size()];
@@ -59,7 +58,7 @@ public static MultiDeviceIterator create(Scope scope, List<String> devices, Stri
     opBuilder.setAttr("devices", devicesArray);
     opBuilder.setAttr("shared_name", sharedName);
     opBuilder.setAttr("container", container);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorFromStringHandle.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorFromStringHandle.java
index a9b0b8ec07e..af6394bb70d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorFromStringHandle.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorFromStringHandle.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -65,11 +64,11 @@ private Options() {
    * @return a new instance of MultiDeviceIteratorFromStringHandle
    */
   @Endpoint(describeByClass = true)
-  public static MultiDeviceIteratorFromStringHandle create(Scope scope, Operand<TString> stringHandle, List<DataType<?>> outputTypes, Options... options) {
+  public static MultiDeviceIteratorFromStringHandle create(Scope scope, Operand<TString> stringHandle, List<Class<? extends TType>> outputTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MultiDeviceIteratorFromStringHandle", scope.makeOpName("MultiDeviceIteratorFromStringHandle"));
     opBuilder.addInput(stringHandle.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorGetNextFromShard.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorGetNextFromShard.java
index 4781bc2d56e..0c0e79f7c31 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorGetNextFromShard.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/MultiDeviceIteratorGetNextFromShard.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,13 +50,13 @@ public final class MultiDeviceIteratorGetNextFromShard extends RawOp implements
    * @return a new instance of MultiDeviceIteratorGetNextFromShard
    */
   @Endpoint(describeByClass = true)
-  public static MultiDeviceIteratorGetNextFromShard create(Scope scope, Operand<?> multiDeviceIterator, Operand<TInt32> shardNum, Operand<TInt64> incarnationId, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static MultiDeviceIteratorGetNextFromShard create(Scope scope, Operand<?> multiDeviceIterator, Operand<TInt32> shardNum, Operand<TInt64> incarnationId, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("MultiDeviceIteratorGetNextFromShard", scope.makeOpName("MultiDeviceIteratorGetNextFromShard"));
     opBuilder.addInput(multiDeviceIterator.asOutput());
     opBuilder.addInput(shardNum.asOutput());
     opBuilder.addInput(incarnationId.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/NonSerializableDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/NonSerializableDataset.java
index 34043a14f9d..52cdcb8ce6f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/NonSerializableDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/NonSerializableDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -44,11 +43,11 @@ public final class NonSerializableDataset extends RawOp implements Operand<TType
    * @return a new instance of NonSerializableDataset
    */
   @Endpoint(describeByClass = true)
-  public static NonSerializableDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static NonSerializableDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("NonSerializableDataset", scope.makeOpName("NonSerializableDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptimizeDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptimizeDataset.java
index 73e27b3ffc9..7025879a980 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptimizeDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptimizeDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -69,12 +68,12 @@ private Options() {
    * @return a new instance of OptimizeDataset
    */
   @Endpoint(describeByClass = true)
-  public static OptimizeDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> optimizations, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static OptimizeDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> optimizations, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OptimizeDataset", scope.makeOpName("OptimizeDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(optimizations.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptionalGetValue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptionalGetValue.java
index 8d3be67da03..97a6bc8aea9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptionalGetValue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/OptionalGetValue.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,11 +47,11 @@ public final class OptionalGetValue extends RawOp implements Iterable<Operand<TT
    * @return a new instance of OptionalGetValue
    */
   @Endpoint(describeByClass = true)
-  public static OptionalGetValue create(Scope scope, Operand<?> optional, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static OptionalGetValue create(Scope scope, Operand<?> optional, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("OptionalGetValue", scope.makeOpName("OptionalGetValue"));
     opBuilder.addInput(optional.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrefetchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrefetchDataset.java
index 78995ad6d20..0da54c538a0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrefetchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrefetchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -77,12 +76,12 @@ private Options() {
    * @return a new instance of PrefetchDataset
    */
   @Endpoint(describeByClass = true)
-  public static PrefetchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> bufferSize, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static PrefetchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> bufferSize, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("PrefetchDataset", scope.makeOpName("PrefetchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(bufferSize.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrivateThreadPoolDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrivateThreadPoolDataset.java
index b06618ce07a..035a6402a74 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrivateThreadPoolDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/PrivateThreadPoolDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class PrivateThreadPoolDataset extends RawOp implements Operand<TTy
    * @return a new instance of PrivateThreadPoolDataset
    */
   @Endpoint(describeByClass = true)
-  public static PrivateThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numThreads, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static PrivateThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numThreads, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("PrivateThreadPoolDataset", scope.makeOpName("PrivateThreadPoolDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numThreads.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RandomDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RandomDataset.java
index 2b944e789dc..577e512c476 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RandomDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RandomDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -60,12 +59,12 @@ public final class RandomDataset extends RawOp implements Operand<TType> {
    * @return a new instance of RandomDataset
    */
   @Endpoint(describeByClass = true)
-  public static RandomDataset create(Scope scope, Operand<TInt64> seed, Operand<TInt64> seed2, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static RandomDataset create(Scope scope, Operand<TInt64> seed, Operand<TInt64> seed2, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("RandomDataset", scope.makeOpName("RandomDataset"));
     opBuilder.addInput(seed.asOutput());
     opBuilder.addInput(seed2.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RangeDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RangeDataset.java
index bc83590e03e..1387e2582e6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RangeDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RangeDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,13 +48,13 @@ public final class RangeDataset extends RawOp implements Operand<TType> {
    * @return a new instance of RangeDataset
    */
   @Endpoint(describeByClass = true)
-  public static RangeDataset create(Scope scope, Operand<TInt64> start, Operand<TInt64> stop, Operand<TInt64> step, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static RangeDataset create(Scope scope, Operand<TInt64> start, Operand<TInt64> stop, Operand<TInt64> step, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("RangeDataset", scope.makeOpName("RangeDataset"));
     opBuilder.addInput(start.asOutput());
     opBuilder.addInput(stop.asOutput());
     opBuilder.addInput(step.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RebatchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RebatchDataset.java
index b5a21c7b6b8..d29ba3dd541 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RebatchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RebatchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -72,12 +71,12 @@ private Options() {
    * @return a new instance of RebatchDataset
    */
   @Endpoint(describeByClass = true)
-  public static RebatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numReplicas, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static RebatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numReplicas, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("RebatchDataset", scope.makeOpName("RebatchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numReplicas.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RepeatDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RepeatDataset.java
index b832025853c..b12001c2439 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RepeatDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/RepeatDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,12 +48,12 @@ public final class RepeatDataset extends RawOp implements Operand<TType> {
    * @return a new instance of RepeatDataset
    */
   @Endpoint(describeByClass = true)
-  public static RepeatDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> count, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static RepeatDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> count, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("RepeatDataset", scope.makeOpName("RepeatDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(count.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SamplingDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SamplingDataset.java
index 7876fe27587..716c2c03128 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SamplingDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SamplingDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -57,14 +56,14 @@ public final class SamplingDataset extends RawOp implements Operand<TType> {
    * @return a new instance of SamplingDataset
    */
   @Endpoint(describeByClass = true)
-  public static SamplingDataset create(Scope scope, Operand<?> inputDataset, Operand<TFloat32> rate, Operand<TInt64> seed, Operand<TInt64> seed2, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SamplingDataset create(Scope scope, Operand<?> inputDataset, Operand<TFloat32> rate, Operand<TInt64> seed, Operand<TInt64> seed2, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("SamplingDataset", scope.makeOpName("SamplingDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(rate.asOutput());
     opBuilder.addInput(seed.asOutput());
     opBuilder.addInput(seed2.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SetStatsAggregatorDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SetStatsAggregatorDataset.java
index 5181c0b1f71..a107e2198b4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SetStatsAggregatorDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SetStatsAggregatorDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,14 +47,14 @@ public final class SetStatsAggregatorDataset extends RawOp implements Operand<TT
    * @return a new instance of SetStatsAggregatorDataset
    */
   @Endpoint(describeByClass = true)
-  public static SetStatsAggregatorDataset create(Scope scope, Operand<?> inputDataset, Operand<?> statsAggregator, Operand<TString> tag, Operand<TString> counterPrefix, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SetStatsAggregatorDataset create(Scope scope, Operand<?> inputDataset, Operand<?> statsAggregator, Operand<TString> tag, Operand<TString> counterPrefix, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("SetStatsAggregatorDataset", scope.makeOpName("SetStatsAggregatorDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(statsAggregator.asOutput());
     opBuilder.addInput(tag.asOutput());
     opBuilder.addInput(counterPrefix.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShardDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShardDataset.java
index 62e18f3fadf..98e043f8e5f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShardDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShardDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -68,13 +67,13 @@ private Options() {
    * @return a new instance of ShardDataset
    */
   @Endpoint(describeByClass = true)
-  public static ShardDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numShards, Operand<TInt64> index, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static ShardDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numShards, Operand<TInt64> index, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ShardDataset", scope.makeOpName("ShardDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numShards.asOutput());
     opBuilder.addInput(index.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleAndRepeatDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleAndRepeatDataset.java
index c5703e8e85c..fa5538deff9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleAndRepeatDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleAndRepeatDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -70,7 +69,7 @@ private Options() {
    * @return a new instance of ShuffleAndRepeatDataset
    */
   @Endpoint(describeByClass = true)
-  public static ShuffleAndRepeatDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> bufferSize, Operand<TInt64> seed, Operand<TInt64> seed2, Operand<TInt64> count, Operand<?> seedGenerator, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static ShuffleAndRepeatDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> bufferSize, Operand<TInt64> seed, Operand<TInt64> seed2, Operand<TInt64> count, Operand<?> seedGenerator, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ShuffleAndRepeatDatasetV2", scope.makeOpName("ShuffleAndRepeatDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(bufferSize.asOutput());
@@ -79,7 +78,7 @@ public static ShuffleAndRepeatDataset create(Scope scope, Operand<?> inputDatase
     opBuilder.addInput(count.asOutput());
     opBuilder.addInput(seedGenerator.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleDataset.java
index 3dd522e319c..e8682dec91c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ShuffleDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -69,7 +68,7 @@ private Options() {
    * @return a new instance of ShuffleDataset
    */
   @Endpoint(describeByClass = true)
-  public static ShuffleDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> bufferSize, Operand<TInt64> seed, Operand<TInt64> seed2, Operand<?> seedGenerator, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static ShuffleDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> bufferSize, Operand<TInt64> seed, Operand<TInt64> seed2, Operand<?> seedGenerator, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ShuffleDatasetV3", scope.makeOpName("ShuffleDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(bufferSize.asOutput());
@@ -77,7 +76,7 @@ public static ShuffleDataset create(Scope scope, Operand<?> inputDataset, Operan
     opBuilder.addInput(seed2.asOutput());
     opBuilder.addInput(seedGenerator.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SkipDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SkipDataset.java
index 4e409b5a4ea..2d7171854a2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SkipDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SkipDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,12 +48,12 @@ public final class SkipDataset extends RawOp implements Operand<TType> {
    * @return a new instance of SkipDataset
    */
   @Endpoint(describeByClass = true)
-  public static SkipDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> count, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SkipDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> count, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("SkipDataset", scope.makeOpName("SkipDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(count.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SleepDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SleepDataset.java
index 67605099106..5feb04bb601 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SleepDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SleepDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class SleepDataset extends RawOp implements Operand<TType> {
    * @return a new instance of SleepDataset
    */
   @Endpoint(describeByClass = true)
-  public static SleepDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> sleepMicroseconds, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SleepDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> sleepMicroseconds, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("SleepDataset", scope.makeOpName("SleepDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(sleepMicroseconds.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SlidingWindowDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SlidingWindowDataset.java
index 89f440efd77..5f4fc56c523 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SlidingWindowDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SlidingWindowDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -52,14 +51,14 @@ public final class SlidingWindowDataset extends RawOp implements Operand<TType>
    * @return a new instance of SlidingWindowDataset
    */
   @Endpoint(describeByClass = true)
-  public static SlidingWindowDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> windowSize, Operand<TInt64> windowShift, Operand<TInt64> windowStride, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SlidingWindowDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> windowSize, Operand<TInt64> windowShift, Operand<TInt64> windowStride, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("SlidingWindowDataset", scope.makeOpName("SlidingWindowDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(windowSize.asOutput());
     opBuilder.addInput(windowShift.asOutput());
     opBuilder.addInput(windowStride.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SnapshotDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SnapshotDataset.java
deleted file mode 100644
index f6bc66e8297..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SnapshotDataset.java
+++ /dev/null
@@ -1,376 +0,0 @@
-/* Copyright 2018 The TensorFlow Authors. All Rights Reserved.
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-=======================================================================*/
-
-// This class has been generated, DO NOT EDIT!
-
-package org.tensorflow.op.data;
-
-import java.util.List;
-import org.tensorflow.DataType;
-import org.tensorflow.Operand;
-import org.tensorflow.Operation;
-import org.tensorflow.OperationBuilder;
-import org.tensorflow.Output;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.op.RawOp;
-import org.tensorflow.op.Scope;
-import org.tensorflow.op.annotation.Endpoint;
-import org.tensorflow.op.annotation.Operator;
-import org.tensorflow.types.TString;
-import org.tensorflow.types.family.TType;
-
-/**
- * Creates a dataset that will write to / read from a snapshot.
- * <p>
- * This dataset attempts to determine whether a valid snapshot exists at the
- * `snapshot_path`, and reads from the snapshot in lieu of using `input_dataset`.
- * If not, it will run the preprocessing pipeline as usual, and write out a
- * snapshot of the data processed for future use.
- */
-public final class SnapshotDataset extends RawOp implements Operand<TType> {
-  
-  /**
-   * Optional attributes for {@link org.tensorflow.op.data.SnapshotDataset}
-   */
-  public static class Options {
-    
-    /**
-     * @param compression 
-     */
-    public Options compression(String compression) {
-      this.compression = compression;
-      return this;
-    }
-    
-    /**
-     * @param readerPathPrefix 
-     */
-    public Options readerPathPrefix(String readerPathPrefix) {
-      this.readerPathPrefix = readerPathPrefix;
-      return this;
-    }
-    
-    /**
-     * @param writerPathPrefix 
-     */
-    public Options writerPathPrefix(String writerPathPrefix) {
-      this.writerPathPrefix = writerPathPrefix;
-      return this;
-    }
-    
-    /**
-     * @param shardSizeBytes 
-     */
-    public Options shardSizeBytes(Long shardSizeBytes) {
-      this.shardSizeBytes = shardSizeBytes;
-      return this;
-    }
-    
-    /**
-     * @param pendingSnapshotExpirySeconds 
-     */
-    public Options pendingSnapshotExpirySeconds(Long pendingSnapshotExpirySeconds) {
-      this.pendingSnapshotExpirySeconds = pendingSnapshotExpirySeconds;
-      return this;
-    }
-    
-    /**
-     * @param numReaderThreads 
-     */
-    public Options numReaderThreads(Long numReaderThreads) {
-      this.numReaderThreads = numReaderThreads;
-      return this;
-    }
-    
-    /**
-     * @param readerBufferSize 
-     */
-    public Options readerBufferSize(Long readerBufferSize) {
-      this.readerBufferSize = readerBufferSize;
-      return this;
-    }
-    
-    /**
-     * @param numWriterThreads 
-     */
-    public Options numWriterThreads(Long numWriterThreads) {
-      this.numWriterThreads = numWriterThreads;
-      return this;
-    }
-    
-    /**
-     * @param writerBufferSize 
-     */
-    public Options writerBufferSize(Long writerBufferSize) {
-      this.writerBufferSize = writerBufferSize;
-      return this;
-    }
-    
-    /**
-     * @param shuffleOnRead 
-     */
-    public Options shuffleOnRead(Boolean shuffleOnRead) {
-      this.shuffleOnRead = shuffleOnRead;
-      return this;
-    }
-    
-    /**
-     * @param seed 
-     */
-    public Options seed(Long seed) {
-      this.seed = seed;
-      return this;
-    }
-    
-    /**
-     * @param seed2 
-     */
-    public Options seed2(Long seed2) {
-      this.seed2 = seed2;
-      return this;
-    }
-    
-    /**
-     * @param mode 
-     */
-    public Options mode(String mode) {
-      this.mode = mode;
-      return this;
-    }
-    
-    /**
-     * @param snapshotName 
-     */
-    public Options snapshotName(String snapshotName) {
-      this.snapshotName = snapshotName;
-      return this;
-    }
-    
-    private String compression;
-    private String readerPathPrefix;
-    private String writerPathPrefix;
-    private Long shardSizeBytes;
-    private Long pendingSnapshotExpirySeconds;
-    private Long numReaderThreads;
-    private Long readerBufferSize;
-    private Long numWriterThreads;
-    private Long writerBufferSize;
-    private Boolean shuffleOnRead;
-    private Long seed;
-    private Long seed2;
-    private String mode;
-    private String snapshotName;
-    
-    private Options() {
-    }
-  }
-  
-  /**
-   * Factory method to create a class wrapping a new SnapshotDataset operation.
-   * 
-   * @param scope current scope
-   * @param inputDataset A variant tensor representing the input dataset.
-   * @param path The path we should write snapshots to / read snapshots from.
-   * @param outputTypes 
-   * @param outputShapes 
-   * @param options carries optional attributes values
-   * @return a new instance of SnapshotDataset
-   */
-  @Endpoint(describeByClass = true)
-  public static SnapshotDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> path, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
-    OperationBuilder opBuilder = scope.env().opBuilder("SnapshotDataset", scope.makeOpName("SnapshotDataset"));
-    opBuilder.addInput(inputDataset.asOutput());
-    opBuilder.addInput(path.asOutput());
-    opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
-    for (int i = 0; i < outputTypesArray.length; ++i) {
-      outputTypesArray[i] = outputTypes.get(i);
-    }
-    opBuilder.setAttr("output_types", outputTypesArray);
-    Shape[] outputShapesArray = new Shape[outputShapes.size()];
-    for (int i = 0; i < outputShapesArray.length; ++i) {
-      outputShapesArray[i] = outputShapes.get(i);
-    }
-    opBuilder.setAttr("output_shapes", outputShapesArray);
-    if (options != null) {
-      for (Options opts : options) {
-        if (opts.compression != null) {
-          opBuilder.setAttr("compression", opts.compression);
-        }
-        if (opts.readerPathPrefix != null) {
-          opBuilder.setAttr("reader_path_prefix", opts.readerPathPrefix);
-        }
-        if (opts.writerPathPrefix != null) {
-          opBuilder.setAttr("writer_path_prefix", opts.writerPathPrefix);
-        }
-        if (opts.shardSizeBytes != null) {
-          opBuilder.setAttr("shard_size_bytes", opts.shardSizeBytes);
-        }
-        if (opts.pendingSnapshotExpirySeconds != null) {
-          opBuilder.setAttr("pending_snapshot_expiry_seconds", opts.pendingSnapshotExpirySeconds);
-        }
-        if (opts.numReaderThreads != null) {
-          opBuilder.setAttr("num_reader_threads", opts.numReaderThreads);
-        }
-        if (opts.readerBufferSize != null) {
-          opBuilder.setAttr("reader_buffer_size", opts.readerBufferSize);
-        }
-        if (opts.numWriterThreads != null) {
-          opBuilder.setAttr("num_writer_threads", opts.numWriterThreads);
-        }
-        if (opts.writerBufferSize != null) {
-          opBuilder.setAttr("writer_buffer_size", opts.writerBufferSize);
-        }
-        if (opts.shuffleOnRead != null) {
-          opBuilder.setAttr("shuffle_on_read", opts.shuffleOnRead);
-        }
-        if (opts.seed != null) {
-          opBuilder.setAttr("seed", opts.seed);
-        }
-        if (opts.seed2 != null) {
-          opBuilder.setAttr("seed2", opts.seed2);
-        }
-        if (opts.mode != null) {
-          opBuilder.setAttr("mode", opts.mode);
-        }
-        if (opts.snapshotName != null) {
-          opBuilder.setAttr("snapshot_name", opts.snapshotName);
-        }
-      }
-    }
-    return new SnapshotDataset(opBuilder.build());
-  }
-  
-  /**
-   * @param compression 
-   */
-  public static Options compression(String compression) {
-    return new Options().compression(compression);
-  }
-  
-  /**
-   * @param readerPathPrefix 
-   */
-  public static Options readerPathPrefix(String readerPathPrefix) {
-    return new Options().readerPathPrefix(readerPathPrefix);
-  }
-  
-  /**
-   * @param writerPathPrefix 
-   */
-  public static Options writerPathPrefix(String writerPathPrefix) {
-    return new Options().writerPathPrefix(writerPathPrefix);
-  }
-  
-  /**
-   * @param shardSizeBytes 
-   */
-  public static Options shardSizeBytes(Long shardSizeBytes) {
-    return new Options().shardSizeBytes(shardSizeBytes);
-  }
-  
-  /**
-   * @param pendingSnapshotExpirySeconds 
-   */
-  public static Options pendingSnapshotExpirySeconds(Long pendingSnapshotExpirySeconds) {
-    return new Options().pendingSnapshotExpirySeconds(pendingSnapshotExpirySeconds);
-  }
-  
-  /**
-   * @param numReaderThreads 
-   */
-  public static Options numReaderThreads(Long numReaderThreads) {
-    return new Options().numReaderThreads(numReaderThreads);
-  }
-  
-  /**
-   * @param readerBufferSize 
-   */
-  public static Options readerBufferSize(Long readerBufferSize) {
-    return new Options().readerBufferSize(readerBufferSize);
-  }
-  
-  /**
-   * @param numWriterThreads 
-   */
-  public static Options numWriterThreads(Long numWriterThreads) {
-    return new Options().numWriterThreads(numWriterThreads);
-  }
-  
-  /**
-   * @param writerBufferSize 
-   */
-  public static Options writerBufferSize(Long writerBufferSize) {
-    return new Options().writerBufferSize(writerBufferSize);
-  }
-  
-  /**
-   * @param shuffleOnRead 
-   */
-  public static Options shuffleOnRead(Boolean shuffleOnRead) {
-    return new Options().shuffleOnRead(shuffleOnRead);
-  }
-  
-  /**
-   * @param seed 
-   */
-  public static Options seed(Long seed) {
-    return new Options().seed(seed);
-  }
-  
-  /**
-   * @param seed2 
-   */
-  public static Options seed2(Long seed2) {
-    return new Options().seed2(seed2);
-  }
-  
-  /**
-   * @param mode 
-   */
-  public static Options mode(String mode) {
-    return new Options().mode(mode);
-  }
-  
-  /**
-   * @param snapshotName 
-   */
-  public static Options snapshotName(String snapshotName) {
-    return new Options().snapshotName(snapshotName);
-  }
-  
-  /**
-   */
-  public Output<?> handle() {
-    return handle;
-  }
-  
-  @Override
-  @SuppressWarnings("unchecked")
-  public Output<TType> asOutput() {
-    return (Output<TType>) handle;
-  }
-  
-  /** The name of this op, as known by TensorFlow core engine */
-  public static final String OP_NAME = "SnapshotDataset";
-  
-  private Output<?> handle;
-  
-  private SnapshotDataset(Operation operation) {
-    super(operation);
-    int outputIdx = 0;
-    handle = operation.output(outputIdx++);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SqlDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SqlDataset.java
index 59a0a47cd4d..cb7140de450 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SqlDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/SqlDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,13 +47,13 @@ public final class SqlDataset extends RawOp implements Operand<TType> {
    * @return a new instance of SqlDataset
    */
   @Endpoint(describeByClass = true)
-  public static SqlDataset create(Scope scope, Operand<TString> driverName, Operand<TString> dataSourceName, Operand<TString> query, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SqlDataset create(Scope scope, Operand<TString> driverName, Operand<TString> dataSourceName, Operand<TString> query, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("SqlDataset", scope.makeOpName("SqlDataset"));
     opBuilder.addInput(driverName.asOutput());
     opBuilder.addInput(dataSourceName.asOutput());
     opBuilder.addInput(query.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/TakeDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/TakeDataset.java
index 3229f055924..8c12d56db8a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/TakeDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/TakeDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -50,12 +49,12 @@ public final class TakeDataset extends RawOp implements Operand<TType> {
    * @return a new instance of TakeDataset
    */
   @Endpoint(describeByClass = true)
-  public static TakeDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> count, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static TakeDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> count, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("TakeDataset", scope.makeOpName("TakeDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(count.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ThreadPoolDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ThreadPoolDataset.java
index 0b1ab263ffa..9cc2ad12bba 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ThreadPoolDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ThreadPoolDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class ThreadPoolDataset extends RawOp implements Operand<TType> {
    * @return a new instance of ThreadPoolDataset
    */
   @Endpoint(describeByClass = true)
-  public static ThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<?> threadPool, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static ThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<?> threadPool, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ThreadPoolDataset", scope.makeOpName("ThreadPoolDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(threadPool.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UnbatchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UnbatchDataset.java
index 1bca9b693cc..0ead7ff2ef1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UnbatchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UnbatchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class UnbatchDataset extends RawOp implements Operand<TType> {
    * @return a new instance of UnbatchDataset
    */
   @Endpoint(describeByClass = true)
-  public static UnbatchDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static UnbatchDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("UnbatchDataset", scope.makeOpName("UnbatchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UniqueDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UniqueDataset.java
index 817519ef6f8..f47548f8dad 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UniqueDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/UniqueDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class UniqueDataset extends RawOp implements Operand<TType> {
    * @return a new instance of UniqueDataset
    */
   @Endpoint(describeByClass = true)
-  public static UniqueDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static UniqueDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("UniqueDataset", scope.makeOpName("UniqueDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/WindowDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/WindowDataset.java
index d5edcc917d8..332338991f5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/WindowDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/WindowDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -98,7 +97,7 @@ public final class WindowDataset extends RawOp implements Operand<TType> {
    * @return a new instance of WindowDataset
    */
   @Endpoint(describeByClass = true)
-  public static WindowDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> size, Operand<TInt64> shift, Operand<TInt64> stride, Operand<TBool> dropRemainder, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static WindowDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> size, Operand<TInt64> shift, Operand<TInt64> stride, Operand<TBool> dropRemainder, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("WindowDataset", scope.makeOpName("WindowDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(size.asOutput());
@@ -106,7 +105,7 @@ public static WindowDataset create(Scope scope, Operand<?> inputDataset, Operand
     opBuilder.addInput(stride.asOutput());
     opBuilder.addInput(dropRemainder.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ZipDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ZipDataset.java
index d4f54946f5f..d5414b73ecf 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ZipDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/ZipDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -53,11 +52,11 @@ public final class ZipDataset extends RawOp implements Operand<TType> {
    * @return a new instance of ZipDataset
    */
   @Endpoint(describeByClass = true)
-  public static ZipDataset create(Scope scope, Iterable<Operand<?>> inputDatasets, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static ZipDataset create(Scope scope, Iterable<Operand<?>> inputDatasets, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ZipDataset", scope.makeOpName("ZipDataset"));
     opBuilder.addInputList(Operands.asOutputs(inputDatasets));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertCardinalityDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertCardinalityDataset.java
index 96c02bf5fc1..6e3694983d4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertCardinalityDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertCardinalityDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class AssertCardinalityDataset extends RawOp implements Operand<TTy
    * @return a new instance of AssertCardinalityDataset
    */
   @Endpoint(describeByClass = true)
-  public static AssertCardinalityDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> cardinality, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static AssertCardinalityDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> cardinality, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("AssertCardinalityDataset", scope.makeOpName("AssertCardinalityDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(cardinality.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertNextDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertNextDataset.java
index cd0c5300df6..f4f0b763dc5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertNextDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AssertNextDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class AssertNextDataset extends RawOp implements Operand<TType> {
    * @return a new instance of AssertNextDataset
    */
   @Endpoint(describeByClass = true)
-  public static AssertNextDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> transformations, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static AssertNextDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> transformations, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalAssertNextDataset", scope.makeOpName("AssertNextDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(transformations.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AutoShardDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AutoShardDataset.java
index 3c1eb053091..d818cdd4175 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AutoShardDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/AutoShardDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -76,13 +75,13 @@ private Options() {
    * @return a new instance of AutoShardDataset
    */
   @Endpoint(describeByClass = true)
-  public static AutoShardDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numWorkers, Operand<TInt64> index, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static AutoShardDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numWorkers, Operand<TInt64> index, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalAutoShardDataset", scope.makeOpName("AutoShardDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numWorkers.asOutput());
     opBuilder.addInput(index.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/BytesProducedStatsDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/BytesProducedStatsDataset.java
index 7fe1290b130..dcf5179b0c9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/BytesProducedStatsDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/BytesProducedStatsDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class BytesProducedStatsDataset extends RawOp implements Operand<TT
    * @return a new instance of BytesProducedStatsDataset
    */
   @Endpoint(describeByClass = true)
-  public static BytesProducedStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static BytesProducedStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalBytesProducedStatsDataset", scope.makeOpName("BytesProducedStatsDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(tag.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ChooseFastestDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ChooseFastestDataset.java
index f731d948490..aee0c946f14 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ChooseFastestDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ChooseFastestDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class ChooseFastestDataset extends RawOp implements Operand<TType>
    * @return a new instance of ChooseFastestDataset
    */
   @Endpoint(describeByClass = true)
-  public static ChooseFastestDataset create(Scope scope, Iterable<Operand<?>> inputDatasets, Long numExperiments, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static ChooseFastestDataset create(Scope scope, Iterable<Operand<?>> inputDatasets, Long numExperiments, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalChooseFastestDataset", scope.makeOpName("ChooseFastestDataset"));
     opBuilder.addInputList(Operands.asOutputs(inputDatasets));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("num_experiments", numExperiments);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DataServiceDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DataServiceDataset.java
index b3e853f9de4..3330b2163e3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DataServiceDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DataServiceDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -73,7 +72,7 @@ private Options() {
    * @return a new instance of DataServiceDataset
    */
   @Endpoint(describeByClass = true)
-  public static DataServiceDataset create(Scope scope, Operand<TInt64> datasetId, Operand<TString> processingMode, Operand<TString> address, Operand<TString> protocol, Operand<TString> jobName, Operand<TInt64> maxOutstandingRequests, Operand<?> iterationCounter, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static DataServiceDataset create(Scope scope, Operand<TInt64> datasetId, Operand<TString> processingMode, Operand<TString> address, Operand<TString> protocol, Operand<TString> jobName, Operand<TInt64> maxOutstandingRequests, Operand<?> iterationCounter, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("DataServiceDataset", scope.makeOpName("DataServiceDataset"));
     opBuilder.addInput(datasetId.asOutput());
     opBuilder.addInput(processingMode.asOutput());
@@ -83,7 +82,7 @@ public static DataServiceDataset create(Scope scope, Operand<TInt64> datasetId,
     opBuilder.addInput(maxOutstandingRequests.asOutput());
     opBuilder.addInput(iterationCounter.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DenseToSparseBatchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DenseToSparseBatchDataset.java
index d464f422836..ff2910a2e38 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DenseToSparseBatchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DenseToSparseBatchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,13 +50,13 @@ public final class DenseToSparseBatchDataset extends RawOp implements Operand<TT
    * @return a new instance of DenseToSparseBatchDataset
    */
   @Endpoint(describeByClass = true)
-  public static DenseToSparseBatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> batchSize, Operand<TInt64> rowShape, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static DenseToSparseBatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> batchSize, Operand<TInt64> rowShape, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalDenseToSparseBatchDataset", scope.makeOpName("DenseToSparseBatchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(batchSize.asOutput());
     opBuilder.addInput(rowShape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DirectedInterleaveDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DirectedInterleaveDataset.java
index 63a06a16201..50ee8a37fcb 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DirectedInterleaveDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/DirectedInterleaveDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,12 +48,12 @@ public final class DirectedInterleaveDataset extends RawOp implements Operand<TT
    * @return a new instance of DirectedInterleaveDataset
    */
   @Endpoint(describeByClass = true)
-  public static DirectedInterleaveDataset create(Scope scope, Operand<?> selectorInputDataset, Iterable<Operand<?>> dataInputDatasets, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static DirectedInterleaveDataset create(Scope scope, Operand<?> selectorInputDataset, Iterable<Operand<?>> dataInputDatasets, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalDirectedInterleaveDataset", scope.makeOpName("DirectedInterleaveDataset"));
     opBuilder.addInput(selectorInputDataset.asOutput());
     opBuilder.addInputList(Operands.asOutputs(dataInputDatasets));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/IgnoreErrorsDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/IgnoreErrorsDataset.java
index 22d734831da..524814790f5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/IgnoreErrorsDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/IgnoreErrorsDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class IgnoreErrorsDataset extends RawOp implements Operand<TType> {
    * @return a new instance of IgnoreErrorsDataset
    */
   @Endpoint(describeByClass = true)
-  public static IgnoreErrorsDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static IgnoreErrorsDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalIgnoreErrorsDataset", scope.makeOpName("IgnoreErrorsDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LatencyStatsDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LatencyStatsDataset.java
index ad901295c28..cd9f0a2c498 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LatencyStatsDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LatencyStatsDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class LatencyStatsDataset extends RawOp implements Operand<TType> {
    * @return a new instance of LatencyStatsDataset
    */
   @Endpoint(describeByClass = true)
-  public static LatencyStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static LatencyStatsDataset create(Scope scope, Operand<?> inputDataset, Operand<TString> tag, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalLatencyStatsDataset", scope.makeOpName("LatencyStatsDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(tag.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LmdbDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LmdbDataset.java
index e6c489833af..3ec97c34b61 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LmdbDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/LmdbDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class LmdbDataset extends RawOp implements Operand<TType> {
    * @return a new instance of LmdbDataset
    */
   @Endpoint(describeByClass = true)
-  public static LmdbDataset create(Scope scope, Operand<TString> filenames, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static LmdbDataset create(Scope scope, Operand<TString> filenames, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalLMDBDataset", scope.makeOpName("LmdbDataset"));
     opBuilder.addInput(filenames.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/MaxIntraOpParallelismDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/MaxIntraOpParallelismDataset.java
index fb69729fab0..88d924e5e43 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/MaxIntraOpParallelismDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/MaxIntraOpParallelismDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class MaxIntraOpParallelismDataset extends RawOp implements Operand
    * @return a new instance of MaxIntraOpParallelismDataset
    */
   @Endpoint(describeByClass = true)
-  public static MaxIntraOpParallelismDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> maxIntraOpParallelism, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static MaxIntraOpParallelismDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> maxIntraOpParallelism, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalMaxIntraOpParallelismDataset", scope.makeOpName("MaxIntraOpParallelismDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(maxIntraOpParallelism.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/NonSerializableDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/NonSerializableDataset.java
index 9f07f37f804..45866ff08b4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/NonSerializableDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/NonSerializableDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -44,11 +43,11 @@ public final class NonSerializableDataset extends RawOp implements Operand<TType
    * @return a new instance of NonSerializableDataset
    */
   @Endpoint(describeByClass = true)
-  public static NonSerializableDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static NonSerializableDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalNonSerializableDataset", scope.makeOpName("NonSerializableDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ParseExampleDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ParseExampleDataset.java
index 4c24cf99558..c4602914a42 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ParseExampleDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ParseExampleDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -102,7 +101,7 @@ private Options() {
    * @return a new instance of ParseExampleDataset
    */
   @Endpoint(describeByClass = true)
-  public static ParseExampleDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numParallelCalls, Iterable<Operand<?>> denseDefaults, List<String> sparseKeys, List<String> denseKeys, List<DataType<?>> sparseTypes, List<Shape> denseShapes, List<DataType<?>> outputTypes, List<Shape> outputShapes, List<DataType<?>> raggedValueTypes, List<DataType<?>> raggedSplitTypes, Options... options) {
+  public static ParseExampleDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numParallelCalls, Iterable<Operand<?>> denseDefaults, List<String> sparseKeys, List<String> denseKeys, List<Class<? extends TType>> sparseTypes, List<Shape> denseShapes, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, List<Class<? extends TType>> raggedValueTypes, List<Class<? extends TNumber>> raggedSplitTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ParseExampleDatasetV2", scope.makeOpName("ParseExampleDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numParallelCalls.asOutput());
@@ -118,7 +117,7 @@ public static ParseExampleDataset create(Scope scope, Operand<?> inputDataset, O
       denseKeysArray[i] = denseKeys.get(i);
     }
     opBuilder.setAttr("dense_keys", denseKeysArray);
-    DataType[] sparseTypesArray = new DataType[sparseTypes.size()];
+    Class[] sparseTypesArray = new Class[sparseTypes.size()];
     for (int i = 0; i < sparseTypesArray.length; ++i) {
       sparseTypesArray[i] = sparseTypes.get(i);
     }
@@ -128,7 +127,7 @@ public static ParseExampleDataset create(Scope scope, Operand<?> inputDataset, O
       denseShapesArray[i] = denseShapes.get(i);
     }
     opBuilder.setAttr("dense_shapes", denseShapesArray);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
@@ -138,12 +137,12 @@ public static ParseExampleDataset create(Scope scope, Operand<?> inputDataset, O
       outputShapesArray[i] = outputShapes.get(i);
     }
     opBuilder.setAttr("output_shapes", outputShapesArray);
-    DataType[] raggedValueTypesArray = new DataType[raggedValueTypes.size()];
+    Class[] raggedValueTypesArray = new Class[raggedValueTypes.size()];
     for (int i = 0; i < raggedValueTypesArray.length; ++i) {
       raggedValueTypesArray[i] = raggedValueTypes.get(i);
     }
     opBuilder.setAttr("ragged_value_types", raggedValueTypesArray);
-    DataType[] raggedSplitTypesArray = new DataType[raggedSplitTypes.size()];
+    Class[] raggedSplitTypesArray = new Class[raggedSplitTypes.size()];
     for (int i = 0; i < raggedSplitTypesArray.length; ++i) {
       raggedSplitTypesArray[i] = raggedSplitTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/PrivateThreadPoolDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/PrivateThreadPoolDataset.java
index a52fd41140e..7d85832bc3e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/PrivateThreadPoolDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/PrivateThreadPoolDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,12 +46,12 @@ public final class PrivateThreadPoolDataset extends RawOp implements Operand<TTy
    * @return a new instance of PrivateThreadPoolDataset
    */
   @Endpoint(describeByClass = true)
-  public static PrivateThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numThreads, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static PrivateThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numThreads, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalPrivateThreadPoolDataset", scope.makeOpName("PrivateThreadPoolDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numThreads.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RandomDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RandomDataset.java
index 8f9109ab3d5..f2e06ee1b97 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RandomDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RandomDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,12 +48,12 @@ public final class RandomDataset extends RawOp implements Operand<TType> {
    * @return a new instance of RandomDataset
    */
   @Endpoint(describeByClass = true)
-  public static RandomDataset create(Scope scope, Operand<TInt64> seed, Operand<TInt64> seed2, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static RandomDataset create(Scope scope, Operand<TInt64> seed, Operand<TInt64> seed2, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalRandomDataset", scope.makeOpName("RandomDataset"));
     opBuilder.addInput(seed.asOutput());
     opBuilder.addInput(seed2.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RebatchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RebatchDataset.java
index 6111166128e..c647ea32f5a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RebatchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/RebatchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -72,12 +71,12 @@ private Options() {
    * @return a new instance of RebatchDataset
    */
   @Endpoint(describeByClass = true)
-  public static RebatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numReplicas, List<DataType<?>> outputTypes, List<Shape> outputShapes, Options... options) {
+  public static RebatchDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> numReplicas, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalRebatchDataset", scope.makeOpName("RebatchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(numReplicas.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SetStatsAggregatorDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SetStatsAggregatorDataset.java
index e2535dcecd7..842c2fae29b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SetStatsAggregatorDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SetStatsAggregatorDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,14 +47,14 @@ public final class SetStatsAggregatorDataset extends RawOp implements Operand<TT
    * @return a new instance of SetStatsAggregatorDataset
    */
   @Endpoint(describeByClass = true)
-  public static SetStatsAggregatorDataset create(Scope scope, Operand<?> inputDataset, Operand<?> statsAggregator, Operand<TString> tag, Operand<TString> counterPrefix, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SetStatsAggregatorDataset create(Scope scope, Operand<?> inputDataset, Operand<?> statsAggregator, Operand<TString> tag, Operand<TString> counterPrefix, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalSetStatsAggregatorDataset", scope.makeOpName("SetStatsAggregatorDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(statsAggregator.asOutput());
     opBuilder.addInput(tag.asOutput());
     opBuilder.addInput(counterPrefix.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SleepDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SleepDataset.java
index 97ddeb96999..cfe2babce9e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SleepDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SleepDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class SleepDataset extends RawOp implements Operand<TType> {
    * @return a new instance of SleepDataset
    */
   @Endpoint(describeByClass = true)
-  public static SleepDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> sleepMicroseconds, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SleepDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> sleepMicroseconds, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalSleepDataset", scope.makeOpName("SleepDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(sleepMicroseconds.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SlidingWindowDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SlidingWindowDataset.java
index b8dbded85e9..a7183757b4f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SlidingWindowDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SlidingWindowDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -52,14 +51,14 @@ public final class SlidingWindowDataset extends RawOp implements Operand<TType>
    * @return a new instance of SlidingWindowDataset
    */
   @Endpoint(describeByClass = true)
-  public static SlidingWindowDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> windowSize, Operand<TInt64> windowShift, Operand<TInt64> windowStride, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SlidingWindowDataset create(Scope scope, Operand<?> inputDataset, Operand<TInt64> windowSize, Operand<TInt64> windowShift, Operand<TInt64> windowStride, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalSlidingWindowDataset", scope.makeOpName("SlidingWindowDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(windowSize.asOutput());
     opBuilder.addInput(windowShift.asOutput());
     opBuilder.addInput(windowStride.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SqlDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SqlDataset.java
index a68f46a5c08..c6bd7b4a433 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SqlDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/SqlDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,13 +47,13 @@ public final class SqlDataset extends RawOp implements Operand<TType> {
    * @return a new instance of SqlDataset
    */
   @Endpoint(describeByClass = true)
-  public static SqlDataset create(Scope scope, Operand<TString> driverName, Operand<TString> dataSourceName, Operand<TString> query, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static SqlDataset create(Scope scope, Operand<TString> driverName, Operand<TString> dataSourceName, Operand<TString> query, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalSqlDataset", scope.makeOpName("SqlDataset"));
     opBuilder.addInput(driverName.asOutput());
     opBuilder.addInput(dataSourceName.asOutput());
     opBuilder.addInput(query.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ThreadPoolDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ThreadPoolDataset.java
index e3fdc2b92cf..2d9a37dba6e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ThreadPoolDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/ThreadPoolDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,12 +45,12 @@ public final class ThreadPoolDataset extends RawOp implements Operand<TType> {
    * @return a new instance of ThreadPoolDataset
    */
   @Endpoint(describeByClass = true)
-  public static ThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<?> threadPool, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static ThreadPoolDataset create(Scope scope, Operand<?> inputDataset, Operand<?> threadPool, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalThreadPoolDataset", scope.makeOpName("ThreadPoolDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder.addInput(threadPool.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UnbatchDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UnbatchDataset.java
index dfa611faa0e..f44f8f09fc2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UnbatchDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UnbatchDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class UnbatchDataset extends RawOp implements Operand<TType> {
    * @return a new instance of UnbatchDataset
    */
   @Endpoint(describeByClass = true)
-  public static UnbatchDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static UnbatchDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalUnbatchDataset", scope.makeOpName("UnbatchDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UncompressElement.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UncompressElement.java
index c5732154e94..79d4624779d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UncompressElement.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UncompressElement.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,11 +46,11 @@ public final class UncompressElement extends RawOp implements Iterable<Operand<T
    * @return a new instance of UncompressElement
    */
   @Endpoint(describeByClass = true)
-  public static UncompressElement create(Scope scope, Operand<?> compressed, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static UncompressElement create(Scope scope, Operand<?> compressed, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("UncompressElement", scope.makeOpName("UncompressElement"));
     opBuilder.addInput(compressed.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UniqueDataset.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UniqueDataset.java
index eb213348e72..b05203c4542 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UniqueDataset.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/data/experimental/UniqueDataset.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.data.experimental;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,11 +44,11 @@ public final class UniqueDataset extends RawOp implements Operand<TType> {
    * @return a new instance of UniqueDataset
    */
   @Endpoint(describeByClass = true)
-  public static UniqueDataset create(Scope scope, Operand<?> inputDataset, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+  public static UniqueDataset create(Scope scope, Operand<?> inputDataset, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExperimentalUniqueDataset", scope.makeOpName("UniqueDataset"));
     opBuilder.addInput(inputDataset.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] outputTypesArray = new DataType[outputTypes.size()];
+    Class[] outputTypesArray = new Class[outputTypes.size()];
     for (int i = 0; i < outputTypesArray.length; ++i) {
       outputTypesArray[i] = outputTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/CheckNumerics.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/CheckNumerics.java
index e26417c561d..44ec8ca8aaf 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/CheckNumerics.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/CheckNumerics.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Checks a tensor for NaN, -Inf and +Inf values.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/DebugNumericsSummary.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/DebugNumericsSummary.java
index 85c20b1eef0..3b927a32cf6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/DebugNumericsSummary.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/debugging/DebugNumericsSummary.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.debugging;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -132,7 +131,7 @@ private Options() {
    * @return a new instance of DebugNumericsSummary
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> DebugNumericsSummary<U> create(Scope scope, Operand<T> input, DataType<U> outputDtype, Options... options) {
+  public static <U extends TNumber, T extends TType> DebugNumericsSummary<U> create(Scope scope, Operand<T> input, Class<U> outputDtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("DebugNumericSummaryV2", scope.makeOpName("DebugNumericsSummary"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -160,7 +159,7 @@ public static <U extends TNumber, T extends TType> DebugNumericsSummary<U> creat
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> DebugNumericsSummary<TFloat32> create(Scope scope, Operand<T> input, Options... options) {
-    return create(scope, input, TFloat32.DTYPE, options);
+    return create(scope, input, TFloat32.class, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Cast.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Cast.java
index 49699f07c8c..66c97240952 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Cast.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Cast.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.dtypes;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -65,7 +64,7 @@ private Options() {
    * @return a new instance of Cast
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> Cast<U> create(Scope scope, Operand<T> x, DataType<U> DstT, Options... options) {
+  public static <U extends TType, T extends TType> Cast<U> create(Scope scope, Operand<T> x, Class<U> DstT, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Cast", scope.makeOpName("Cast"));
     opBuilder.addInput(x.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Complex.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Complex.java
index 5e8918785a0..4d245490516 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Complex.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/dtypes/Complex.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.dtypes;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -62,7 +61,7 @@ public final class Complex<U extends TType> extends RawOp implements Operand<U>
    * @return a new instance of Complex
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TNumber> Complex<U> create(Scope scope, Operand<T> real, Operand<T> imag, DataType<U> Tout) {
+  public static <U extends TType, T extends TNumber> Complex<U> create(Scope scope, Operand<T> real, Operand<T> imag, Class<U> Tout) {
     OperationBuilder opBuilder = scope.env().opBuilder("Complex", scope.makeOpName("Complex"));
     opBuilder.addInput(real.asOutput());
     opBuilder.addInput(imag.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustContrast.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustContrast.java
index ab5b3f32e61..b33b91ed953 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustContrast.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustContrast.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Adjust the contrast of one or more images.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustHue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustHue.java
index 2fe15aa50f1..7f1c320f052 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustHue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustHue.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Adjust the hue of one or more images.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustSaturation.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustSaturation.java
index 03270949652..49a7f6ef6ae 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustSaturation.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/AdjustSaturation.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Adjust the saturation of one or more images.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResize.java
index f0048c1014b..58e9ed905d1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResize.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Extracts crops from the input image tensor and resizes them.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradBoxes.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradBoxes.java
index 67263c1e576..4c3c0f678c1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradBoxes.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradBoxes.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of the crop_and_resize op wrt the input boxes tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradImage.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradImage.java
index 6a6415f879d..32a7bc6d247 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradImage.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/CropAndResizeGradImage.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.image;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of the crop_and_resize op wrt the input image tensor.
@@ -84,7 +82,7 @@ private Options() {
    * @return a new instance of CropAndResizeGradImage
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> CropAndResizeGradImage<T> create(Scope scope, Operand<TFloat32> grads, Operand<TFloat32> boxes, Operand<TInt32> boxInd, Operand<TInt32> imageSize, DataType<T> T, Options... options) {
+  public static <T extends TNumber> CropAndResizeGradImage<T> create(Scope scope, Operand<TFloat32> grads, Operand<TFloat32> boxes, Operand<TInt32> boxInd, Operand<TInt32> imageSize, Class<T> T, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("CropAndResizeGradImage", scope.makeOpName("CropAndResizeGradImage"));
     opBuilder.addInput(grads.asOutput());
     opBuilder.addInput(boxes.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DecodePng.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DecodePng.java
index 1d97bc54df7..dc917cb869a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DecodePng.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DecodePng.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.image;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TString;
 import org.tensorflow.types.TUint8;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Decode a PNG-encoded image to a uint8 or uint16 tensor.
@@ -92,7 +90,7 @@ private Options() {
    * @return a new instance of DecodePng
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> DecodePng<T> create(Scope scope, Operand<TString> contents, DataType<T> dtype, Options... options) {
+  public static <T extends TNumber> DecodePng<T> create(Scope scope, Operand<TString> contents, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("DecodePng", scope.makeOpName("DecodePng"));
     opBuilder.addInput(contents.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -117,7 +115,7 @@ public static <T extends TNumber> DecodePng<T> create(Scope scope, Operand<TStri
    */
   @Endpoint(describeByClass = true)
   public static DecodePng<TUint8> create(Scope scope, Operand<TString> contents, Options... options) {
-    return create(scope, contents, TUint8.DTYPE, options);
+    return create(scope, contents, TUint8.class, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DrawBoundingBoxes.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DrawBoundingBoxes.java
index ba674376d07..edd3f0654d0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DrawBoundingBoxes.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/DrawBoundingBoxes.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Draw bounding boxes on a batch of images.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/EncodePng.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/EncodePng.java
index cf5b5a0ebc5..5944fbc9f87 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/EncodePng.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/EncodePng.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * PNG-encode an image.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ExtractJpegShape.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ExtractJpegShape.java
index dfa8c4ea7b3..1f5ff7163fb 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ExtractJpegShape.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ExtractJpegShape.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.image;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Extract the shape information of a JPEG-encoded image.
@@ -51,7 +49,7 @@ public final class ExtractJpegShape<T extends TNumber> extends RawOp implements
    * @return a new instance of ExtractJpegShape
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> ExtractJpegShape<T> create(Scope scope, Operand<TString> contents, DataType<T> outputType) {
+  public static <T extends TNumber> ExtractJpegShape<T> create(Scope scope, Operand<TString> contents, Class<T> outputType) {
     OperationBuilder opBuilder = scope.env().opBuilder("ExtractJpegShape", scope.makeOpName("ExtractJpegShape"));
     opBuilder.addInput(contents.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -68,7 +66,7 @@ public static <T extends TNumber> ExtractJpegShape<T> create(Scope scope, Operan
    */
   @Endpoint(describeByClass = true)
   public static ExtractJpegShape<TInt32> create(Scope scope, Operand<TString> contents) {
-    return create(scope, contents, TInt32.DTYPE);
+    return create(scope, contents, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/HsvToRgb.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/HsvToRgb.java
index fb752a83518..90b9c5ea776 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/HsvToRgb.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/HsvToRgb.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Convert one or more images from HSV to RGB.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ImageProjectiveTransformV2.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ImageProjectiveTransformV2.java
index 3363d6a9804..a1c786cb619 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ImageProjectiveTransformV2.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ImageProjectiveTransformV2.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Applies the given transform to each of the images.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/NonMaxSuppression.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/NonMaxSuppression.java
index 0745299e34f..919a761f5c4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/NonMaxSuppression.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/NonMaxSuppression.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Greedily selects a subset of bounding boxes in descending order of score,
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RandomCrop.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RandomCrop.java
index 47d03d63187..a5e2de121b1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RandomCrop.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RandomCrop.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Randomly crop `image`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeArea.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeArea.java
index 60d1e473ef0..895bdfac587 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeArea.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeArea.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Resize `images` to `size` using area interpolation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubic.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubic.java
index 2519eaa6ea9..7b3b9ffdcdd 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubic.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubic.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Resize `images` to `size` using bicubic interpolation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubicGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubicGrad.java
index 96296cdf33f..e57cd749a72 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubicGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBicubicGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of bicubic interpolation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinear.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinear.java
index d0c28fed013..8b3bba2c9ae 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinear.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinear.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Resize `images` to `size` using bilinear interpolation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinearGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinearGrad.java
index b8aef51ba30..572f8693d4b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinearGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeBilinearGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of bilinear interpolation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighbor.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighbor.java
index dd7865841de..1da719fb569 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighbor.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighbor.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Resize `images` to `size` using nearest neighbor interpolation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighborGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighborGrad.java
index 4897c0f9dc6..f0139dcca6c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighborGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ResizeNearestNeighborGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of nearest neighbor interpolation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RgbToHsv.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RgbToHsv.java
index acee7a6061e..0ac21df35a5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RgbToHsv.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/RgbToHsv.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Converts one or more images from RGB to HSV.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/SampleDistortedBoundingBox.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/SampleDistortedBoundingBox.java
index 7f83b2ee84c..583da52ce0e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/SampleDistortedBoundingBox.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/SampleDistortedBoundingBox.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Generate a single randomly distorted bounding box for an image.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslate.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslate.java
index 26bf544be2e..cf52ddc87d4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslate.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslate.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  */
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslateGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslateGrad.java
index 4b8d9b0fc4b..79d6c35a46e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslateGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/image/ScaleAndTranslateGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodePaddedRaw.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodePaddedRaw.java
index 820b0077ba5..87a4ed8300c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodePaddedRaw.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodePaddedRaw.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.io;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Reinterpret the bytes of a string as a vector of numbers.
@@ -71,7 +69,7 @@ private Options() {
    * @return a new instance of DecodePaddedRaw
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> DecodePaddedRaw<T> create(Scope scope, Operand<TString> inputBytes, Operand<TInt32> fixedLength, DataType<T> outType, Options... options) {
+  public static <T extends TNumber> DecodePaddedRaw<T> create(Scope scope, Operand<TString> inputBytes, Operand<TInt32> fixedLength, Class<T> outType, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("DecodePaddedRaw", scope.makeOpName("DecodePaddedRaw"));
     opBuilder.addInput(inputBytes.asOutput());
     opBuilder.addInput(fixedLength.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodeRaw.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodeRaw.java
index dfdabfdb466..24b013325f3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodeRaw.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DecodeRaw.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.io;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -68,7 +67,7 @@ private Options() {
    * @return a new instance of DecodeRaw
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> DecodeRaw<T> create(Scope scope, Operand<TString> bytes, DataType<T> outType, Options... options) {
+  public static <T extends TType> DecodeRaw<T> create(Scope scope, Operand<TString> bytes, Class<T> outType, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("DecodeRaw", scope.makeOpName("DecodeRaw"));
     opBuilder.addInput(bytes.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DeserializeManySparse.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DeserializeManySparse.java
index 8582681c073..2a00c03b6a2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DeserializeManySparse.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/DeserializeManySparse.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.io;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -90,7 +89,7 @@ public final class DeserializeManySparse<T extends TType> extends RawOp {
    * @return a new instance of DeserializeManySparse
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> DeserializeManySparse<T> create(Scope scope, Operand<TString> serializedSparse, DataType<T> dtype) {
+  public static <T extends TType> DeserializeManySparse<T> create(Scope scope, Operand<TString> serializedSparse, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("DeserializeManySparse", scope.makeOpName("DeserializeManySparse"));
     opBuilder.addInput(serializedSparse.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/FifoQueue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/FifoQueue.java
index 842985b49e9..0e507d45e89 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/FifoQueue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/FifoQueue.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.io;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -97,10 +96,10 @@ private Options() {
    * @return a new instance of FifoQueue
    */
   @Endpoint(describeByClass = true)
-  public static FifoQueue create(Scope scope, List<DataType<?>> componentTypes, Options... options) {
+  public static FifoQueue create(Scope scope, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("FIFOQueueV2", scope.makeOpName("FifoQueue"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PaddingFifoQueue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PaddingFifoQueue.java
index b72d8188038..9efc68bbdc7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PaddingFifoQueue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PaddingFifoQueue.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.io;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -105,10 +104,10 @@ private Options() {
    * @return a new instance of PaddingFifoQueue
    */
   @Endpoint(describeByClass = true)
-  public static PaddingFifoQueue create(Scope scope, List<DataType<?>> componentTypes, Options... options) {
+  public static PaddingFifoQueue create(Scope scope, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("PaddingFIFOQueueV2", scope.makeOpName("PaddingFifoQueue"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseExample.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseExample.java
index bd8f0f4a603..cbcf7e4deef 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseExample.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseExample.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -33,6 +32,7 @@
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
+import org.tensorflow.types.family.TType;
 
 /**
  * Transforms a vector of tf.Example protos (as strings) into typed tensors.
@@ -97,7 +97,7 @@ public final class ParseExample extends RawOp {
    * @return a new instance of ParseExample
    */
   @Endpoint(describeByClass = true)
-  public static ParseExample create(Scope scope, Operand<TString> serialized, Operand<TString> names, Operand<TString> sparseKeys, Operand<TString> denseKeys, Operand<TString> raggedKeys, Iterable<Operand<?>> denseDefaults, Long numSparse, List<DataType<?>> sparseTypes, List<DataType<?>> raggedValueTypes, List<DataType<?>> raggedSplitTypes, List<Shape> denseShapes) {
+  public static ParseExample create(Scope scope, Operand<TString> serialized, Operand<TString> names, Operand<TString> sparseKeys, Operand<TString> denseKeys, Operand<TString> raggedKeys, Iterable<Operand<?>> denseDefaults, Long numSparse, List<Class<? extends TType>> sparseTypes, List<Class<? extends TType>> raggedValueTypes, List<Class<? extends TNumber>> raggedSplitTypes, List<Shape> denseShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ParseExampleV2", scope.makeOpName("ParseExample"));
     opBuilder.addInput(serialized.asOutput());
     opBuilder.addInput(names.asOutput());
@@ -107,17 +107,17 @@ public static ParseExample create(Scope scope, Operand<TString> serialized, Oper
     opBuilder.addInputList(Operands.asOutputs(denseDefaults));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("num_sparse", numSparse);
-    DataType[] sparseTypesArray = new DataType[sparseTypes.size()];
+    Class[] sparseTypesArray = new Class[sparseTypes.size()];
     for (int i = 0; i < sparseTypesArray.length; ++i) {
       sparseTypesArray[i] = sparseTypes.get(i);
     }
     opBuilder.setAttr("sparse_types", sparseTypesArray);
-    DataType[] raggedValueTypesArray = new DataType[raggedValueTypes.size()];
+    Class[] raggedValueTypesArray = new Class[raggedValueTypes.size()];
     for (int i = 0; i < raggedValueTypesArray.length; ++i) {
       raggedValueTypesArray[i] = raggedValueTypes.get(i);
     }
     opBuilder.setAttr("ragged_value_types", raggedValueTypesArray);
-    DataType[] raggedSplitTypesArray = new DataType[raggedSplitTypes.size()];
+    Class[] raggedSplitTypesArray = new Class[raggedSplitTypes.size()];
     for (int i = 0; i < raggedSplitTypesArray.length; ++i) {
       raggedSplitTypesArray[i] = raggedSplitTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSequenceExample.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSequenceExample.java
index 1c10b138c55..df0db9fabc0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSequenceExample.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSequenceExample.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -34,6 +33,7 @@
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
+import org.tensorflow.types.family.TType;
 
 /**
  * Transforms a vector of tf.io.SequenceExample protos (as strings) into
@@ -153,7 +153,7 @@ private Options() {
    * @return a new instance of ParseSequenceExample
    */
   @Endpoint(describeByClass = true)
-  public static ParseSequenceExample create(Scope scope, Operand<TString> serialized, Operand<TString> debugName, Operand<TString> contextSparseKeys, Operand<TString> contextDenseKeys, Operand<TString> contextRaggedKeys, Operand<TString> featureListSparseKeys, Operand<TString> featureListDenseKeys, Operand<TString> featureListRaggedKeys, Operand<TBool> featureListDenseMissingAssumedEmpty, Iterable<Operand<?>> contextDenseDefaults, List<DataType<?>> contextSparseTypes, List<DataType<?>> contextRaggedValueTypes, List<DataType<?>> contextRaggedSplitTypes, List<DataType<?>> featureListDenseTypes, List<DataType<?>> featureListSparseTypes, List<DataType<?>> featureListRaggedValueTypes, List<DataType<?>> featureListRaggedSplitTypes, Options... options) {
+  public static ParseSequenceExample create(Scope scope, Operand<TString> serialized, Operand<TString> debugName, Operand<TString> contextSparseKeys, Operand<TString> contextDenseKeys, Operand<TString> contextRaggedKeys, Operand<TString> featureListSparseKeys, Operand<TString> featureListDenseKeys, Operand<TString> featureListRaggedKeys, Operand<TBool> featureListDenseMissingAssumedEmpty, Iterable<Operand<?>> contextDenseDefaults, List<Class<? extends TType>> contextSparseTypes, List<Class<? extends TType>> contextRaggedValueTypes, List<Class<? extends TNumber>> contextRaggedSplitTypes, List<Class<? extends TType>> featureListDenseTypes, List<Class<? extends TType>> featureListSparseTypes, List<Class<? extends TType>> featureListRaggedValueTypes, List<Class<? extends TNumber>> featureListRaggedSplitTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ParseSequenceExampleV2", scope.makeOpName("ParseSequenceExample"));
     opBuilder.addInput(serialized.asOutput());
     opBuilder.addInput(debugName.asOutput());
@@ -166,37 +166,37 @@ public static ParseSequenceExample create(Scope scope, Operand<TString> serializ
     opBuilder.addInput(featureListDenseMissingAssumedEmpty.asOutput());
     opBuilder.addInputList(Operands.asOutputs(contextDenseDefaults));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] contextSparseTypesArray = new DataType[contextSparseTypes.size()];
+    Class[] contextSparseTypesArray = new Class[contextSparseTypes.size()];
     for (int i = 0; i < contextSparseTypesArray.length; ++i) {
       contextSparseTypesArray[i] = contextSparseTypes.get(i);
     }
     opBuilder.setAttr("context_sparse_types", contextSparseTypesArray);
-    DataType[] contextRaggedValueTypesArray = new DataType[contextRaggedValueTypes.size()];
+    Class[] contextRaggedValueTypesArray = new Class[contextRaggedValueTypes.size()];
     for (int i = 0; i < contextRaggedValueTypesArray.length; ++i) {
       contextRaggedValueTypesArray[i] = contextRaggedValueTypes.get(i);
     }
     opBuilder.setAttr("context_ragged_value_types", contextRaggedValueTypesArray);
-    DataType[] contextRaggedSplitTypesArray = new DataType[contextRaggedSplitTypes.size()];
+    Class[] contextRaggedSplitTypesArray = new Class[contextRaggedSplitTypes.size()];
     for (int i = 0; i < contextRaggedSplitTypesArray.length; ++i) {
       contextRaggedSplitTypesArray[i] = contextRaggedSplitTypes.get(i);
     }
     opBuilder.setAttr("context_ragged_split_types", contextRaggedSplitTypesArray);
-    DataType[] featureListDenseTypesArray = new DataType[featureListDenseTypes.size()];
+    Class[] featureListDenseTypesArray = new Class[featureListDenseTypes.size()];
     for (int i = 0; i < featureListDenseTypesArray.length; ++i) {
       featureListDenseTypesArray[i] = featureListDenseTypes.get(i);
     }
     opBuilder.setAttr("feature_list_dense_types", featureListDenseTypesArray);
-    DataType[] featureListSparseTypesArray = new DataType[featureListSparseTypes.size()];
+    Class[] featureListSparseTypesArray = new Class[featureListSparseTypes.size()];
     for (int i = 0; i < featureListSparseTypesArray.length; ++i) {
       featureListSparseTypesArray[i] = featureListSparseTypes.get(i);
     }
     opBuilder.setAttr("feature_list_sparse_types", featureListSparseTypesArray);
-    DataType[] featureListRaggedValueTypesArray = new DataType[featureListRaggedValueTypes.size()];
+    Class[] featureListRaggedValueTypesArray = new Class[featureListRaggedValueTypes.size()];
     for (int i = 0; i < featureListRaggedValueTypesArray.length; ++i) {
       featureListRaggedValueTypesArray[i] = featureListRaggedValueTypes.get(i);
     }
     opBuilder.setAttr("feature_list_ragged_value_types", featureListRaggedValueTypesArray);
-    DataType[] featureListRaggedSplitTypesArray = new DataType[featureListRaggedSplitTypes.size()];
+    Class[] featureListRaggedSplitTypesArray = new Class[featureListRaggedSplitTypes.size()];
     for (int i = 0; i < featureListRaggedSplitTypesArray.length; ++i) {
       featureListRaggedSplitTypesArray[i] = featureListRaggedSplitTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleExample.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleExample.java
index 2c09ba8ac4c..597e75efa9e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleExample.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleExample.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -32,6 +31,7 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 /**
  * Transforms a tf.Example proto (as a string) into typed tensors.
@@ -76,7 +76,7 @@ public final class ParseSingleExample extends RawOp {
    * @return a new instance of ParseSingleExample
    */
   @Endpoint(describeByClass = true)
-  public static ParseSingleExample create(Scope scope, Operand<TString> serialized, Iterable<Operand<?>> denseDefaults, Long numSparse, List<String> sparseKeys, List<String> denseKeys, List<DataType<?>> sparseTypes, List<Shape> denseShapes) {
+  public static ParseSingleExample create(Scope scope, Operand<TString> serialized, Iterable<Operand<?>> denseDefaults, Long numSparse, List<String> sparseKeys, List<String> denseKeys, List<Class<? extends TType>> sparseTypes, List<Shape> denseShapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("ParseSingleExample", scope.makeOpName("ParseSingleExample"));
     opBuilder.addInput(serialized.asOutput());
     opBuilder.addInputList(Operands.asOutputs(denseDefaults));
@@ -92,7 +92,7 @@ public static ParseSingleExample create(Scope scope, Operand<TString> serialized
       denseKeysArray[i] = denseKeys.get(i);
     }
     opBuilder.setAttr("dense_keys", denseKeysArray);
-    DataType[] sparseTypesArray = new DataType[sparseTypes.size()];
+    Class[] sparseTypesArray = new Class[sparseTypes.size()];
     for (int i = 0; i < sparseTypesArray.length; ++i) {
       sparseTypesArray[i] = sparseTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleSequenceExample.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleSequenceExample.java
index bb1e1eaa576..89105dc77ab 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleSequenceExample.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseSingleSequenceExample.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -32,6 +31,7 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 /**
  * Transforms a scalar brain.SequenceExample proto (as strings) into typed tensors.
@@ -122,7 +122,7 @@ private Options() {
    * @return a new instance of ParseSingleSequenceExample
    */
   @Endpoint(describeByClass = true)
-  public static ParseSingleSequenceExample create(Scope scope, Operand<TString> serialized, Operand<TString> featureListDenseMissingAssumedEmpty, Iterable<Operand<TString>> contextSparseKeys, Iterable<Operand<TString>> contextDenseKeys, Iterable<Operand<TString>> featureListSparseKeys, Iterable<Operand<TString>> featureListDenseKeys, Iterable<Operand<?>> contextDenseDefaults, Operand<TString> debugName, List<DataType<?>> contextSparseTypes, List<DataType<?>> featureListDenseTypes, List<DataType<?>> featureListSparseTypes, Options... options) {
+  public static ParseSingleSequenceExample create(Scope scope, Operand<TString> serialized, Operand<TString> featureListDenseMissingAssumedEmpty, Iterable<Operand<TString>> contextSparseKeys, Iterable<Operand<TString>> contextDenseKeys, Iterable<Operand<TString>> featureListSparseKeys, Iterable<Operand<TString>> featureListDenseKeys, Iterable<Operand<?>> contextDenseDefaults, Operand<TString> debugName, List<Class<? extends TType>> contextSparseTypes, List<Class<? extends TType>> featureListDenseTypes, List<Class<? extends TType>> featureListSparseTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ParseSingleSequenceExample", scope.makeOpName("ParseSingleSequenceExample"));
     opBuilder.addInput(serialized.asOutput());
     opBuilder.addInput(featureListDenseMissingAssumedEmpty.asOutput());
@@ -133,17 +133,17 @@ public static ParseSingleSequenceExample create(Scope scope, Operand<TString> se
     opBuilder.addInputList(Operands.asOutputs(contextDenseDefaults));
     opBuilder.addInput(debugName.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] contextSparseTypesArray = new DataType[contextSparseTypes.size()];
+    Class[] contextSparseTypesArray = new Class[contextSparseTypes.size()];
     for (int i = 0; i < contextSparseTypesArray.length; ++i) {
       contextSparseTypesArray[i] = contextSparseTypes.get(i);
     }
     opBuilder.setAttr("context_sparse_types", contextSparseTypesArray);
-    DataType[] featureListDenseTypesArray = new DataType[featureListDenseTypes.size()];
+    Class[] featureListDenseTypesArray = new Class[featureListDenseTypes.size()];
     for (int i = 0; i < featureListDenseTypesArray.length; ++i) {
       featureListDenseTypesArray[i] = featureListDenseTypes.get(i);
     }
     opBuilder.setAttr("feature_list_dense_types", featureListDenseTypesArray);
-    DataType[] featureListSparseTypesArray = new DataType[featureListSparseTypes.size()];
+    Class[] featureListSparseTypesArray = new Class[featureListSparseTypes.size()];
     for (int i = 0; i < featureListSparseTypesArray.length; ++i) {
       featureListSparseTypesArray[i] = featureListSparseTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseTensor.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseTensor.java
index bebdc1b4419..a45fa3a542f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseTensor.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/ParseTensor.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.io;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,7 +46,7 @@ public final class ParseTensor<T extends TType> extends RawOp implements Operand
    * @return a new instance of ParseTensor
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> ParseTensor<T> create(Scope scope, Operand<TString> serialized, DataType<T> outType) {
+  public static <T extends TType> ParseTensor<T> create(Scope scope, Operand<TString> serialized, Class<T> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("ParseTensor", scope.makeOpName("ParseTensor"));
     opBuilder.addInput(serialized.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PriorityQueue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PriorityQueue.java
index 569d4ae7eb3..b8aba81d587 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PriorityQueue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/PriorityQueue.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.io;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -95,10 +94,10 @@ private Options() {
    * @return a new instance of PriorityQueue
    */
   @Endpoint(describeByClass = true)
-  public static PriorityQueue create(Scope scope, List<DataType<?>> componentTypes, List<Shape> shapes, Options... options) {
+  public static PriorityQueue create(Scope scope, List<Class<? extends TType>> componentTypes, List<Shape> shapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("PriorityQueueV2", scope.makeOpName("PriorityQueue"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeue.java
index b74fb3ad3e8..bf0cce74ece 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeue.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -75,11 +74,11 @@ private Options() {
    * @return a new instance of QueueDequeue
    */
   @Endpoint(describeByClass = true)
-  public static QueueDequeue create(Scope scope, Operand<?> handle, List<DataType<?>> componentTypes, Options... options) {
+  public static QueueDequeue create(Scope scope, Operand<?> handle, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QueueDequeueV2", scope.makeOpName("QueueDequeue"));
     opBuilder.addInput(handle.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueMany.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueMany.java
index 2906fe6543c..c24cb4251f6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueMany.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueMany.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -84,12 +83,12 @@ private Options() {
    * @return a new instance of QueueDequeueMany
    */
   @Endpoint(describeByClass = true)
-  public static QueueDequeueMany create(Scope scope, Operand<?> handle, Operand<TInt32> n, List<DataType<?>> componentTypes, Options... options) {
+  public static QueueDequeueMany create(Scope scope, Operand<?> handle, Operand<TInt32> n, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QueueDequeueManyV2", scope.makeOpName("QueueDequeueMany"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(n.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueUpTo.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueUpTo.java
index 51af788104c..d2302e77dea 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueUpTo.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/QueueDequeueUpTo.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -88,12 +87,12 @@ private Options() {
    * @return a new instance of QueueDequeueUpTo
    */
   @Endpoint(describeByClass = true)
-  public static QueueDequeueUpTo create(Scope scope, Operand<?> handle, Operand<TInt32> n, List<DataType<?>> componentTypes, Options... options) {
+  public static QueueDequeueUpTo create(Scope scope, Operand<?> handle, Operand<TInt32> n, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QueueDequeueUpToV2", scope.makeOpName("QueueDequeueUpTo"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(n.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/RandomShuffleQueue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/RandomShuffleQueue.java
index cdde1debc95..eb2d6a4ad52 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/RandomShuffleQueue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/RandomShuffleQueue.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.io;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -127,10 +126,10 @@ private Options() {
    * @return a new instance of RandomShuffleQueue
    */
   @Endpoint(describeByClass = true)
-  public static RandomShuffleQueue create(Scope scope, List<DataType<?>> componentTypes, Options... options) {
+  public static RandomShuffleQueue create(Scope scope, List<Class<? extends TType>> componentTypes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("RandomShuffleQueueV2", scope.makeOpName("RandomShuffleQueue"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] componentTypesArray = new DataType[componentTypes.size()];
+    Class[] componentTypesArray = new Class[componentTypes.size()];
     for (int i = 0; i < componentTypesArray.length; ++i) {
       componentTypesArray[i] = componentTypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeManySparse.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeManySparse.java
index 0afbdcd6edd..35b1bbfae09 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeManySparse.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeManySparse.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.io;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -58,7 +57,7 @@ public final class SerializeManySparse<U extends TType> extends RawOp implements
    * @return a new instance of SerializeManySparse
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> SerializeManySparse<U> create(Scope scope, Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape, DataType<U> outType) {
+  public static <U extends TType, T extends TType> SerializeManySparse<U> create(Scope scope, Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("SerializeManySparse", scope.makeOpName("SerializeManySparse"));
     opBuilder.addInput(sparseIndices.asOutput());
     opBuilder.addInput(sparseValues.asOutput());
@@ -79,7 +78,7 @@ public static <U extends TType, T extends TType> SerializeManySparse<U> create(S
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> SerializeManySparse<TString> create(Scope scope, Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape) {
-    return create(scope, sparseIndices, sparseValues, sparseShape, TString.DTYPE);
+    return create(scope, sparseIndices, sparseValues, sparseShape, TString.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeSparse.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeSparse.java
index 2d8fc7838e4..fc4e5c09f21 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeSparse.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/io/SerializeSparse.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.io;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -50,7 +49,7 @@ public final class SerializeSparse<U extends TType> extends RawOp implements Ope
    * @return a new instance of SerializeSparse
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> SerializeSparse<U> create(Scope scope, Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape, DataType<U> outType) {
+  public static <U extends TType, T extends TType> SerializeSparse<U> create(Scope scope, Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("SerializeSparse", scope.makeOpName("SerializeSparse"));
     opBuilder.addInput(sparseIndices.asOutput());
     opBuilder.addInput(sparseValues.asOutput());
@@ -71,7 +70,7 @@ public static <U extends TType, T extends TType> SerializeSparse<U> create(Scope
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> SerializeSparse<TString> create(Scope scope, Operand<TInt64> sparseIndices, Operand<T> sparseValues, Operand<TInt64> sparseShape) {
-    return create(scope, sparseIndices, sparseValues, sparseShape, TString.DTYPE);
+    return create(scope, sparseIndices, sparseValues, sparseShape, TString.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholesky.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholesky.java
index 733139f4359..40a66933f4f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholesky.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholesky.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholeskyGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholeskyGrad.java
index d56f45e8abf..caa625ffe79 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholeskyGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchCholeskyGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixInverse.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixInverse.java
index 36bbee414e9..69017d8dcfc 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixInverse.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixInverse.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolve.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolve.java
index 677b8aa0e7b..2748e30a02b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolve.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolve.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolveLs.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolveLs.java
index ae4a4366b3b..ca30c69f971 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolveLs.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixSolveLs.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixTriangularSolve.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixTriangularSolve.java
index cb1baa772fd..c747840b107 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixTriangularSolve.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchMatrixTriangularSolve.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchSelfAdjointEig.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchSelfAdjointEig.java
index e589f5c49fe..e85fea495fd 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchSelfAdjointEig.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/BatchSelfAdjointEig.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code e()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/CholeskyGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/CholeskyGrad.java
index b4171907853..e5ac38214dc 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/CholeskyGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/CholeskyGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the reverse mode backpropagated gradient of the Cholesky algorithm.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Cross.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Cross.java
index 0afc8a2bf60..d223c04922e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Cross.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Cross.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Compute the pairwise cross product.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Eig.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Eig.java
index 68c33487129..e02e54defe6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Eig.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Eig.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -78,7 +77,7 @@ private Options() {
    * @return a new instance of Eig
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> Eig<U> create(Scope scope, Operand<T> input, DataType<U> Tout, Options... options) {
+  public static <U extends TType, T extends TType> Eig<U> create(Scope scope, Operand<T> input, Class<U> Tout, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Eig", scope.makeOpName("Eig"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Lu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Lu.java
index ce37bbcbea2..fcbdbd6d80d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Lu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/Lu.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -67,7 +66,7 @@ public final class Lu<T extends TType, U extends TNumber> extends RawOp {
    * @return a new instance of Lu
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TNumber> Lu<T, U> create(Scope scope, Operand<T> input, DataType<U> outputIdxType) {
+  public static <T extends TType, U extends TNumber> Lu<T, U> create(Scope scope, Operand<T> input, Class<U> outputIdxType) {
     OperationBuilder opBuilder = scope.env().opBuilder("Lu", scope.makeOpName("Lu"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -85,7 +84,7 @@ public static <T extends TType, U extends TNumber> Lu<T, U> create(Scope scope,
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Lu<T, TInt32> create(Scope scope, Operand<T> input) {
-    return create(scope, input, TInt32.DTYPE);
+    return create(scope, input, TInt32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMul.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMul.java
index 445d27e1591..0fc70a09b7d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMul.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMul.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -87,7 +86,7 @@ private Options() {
    * @return a new instance of QuantizedMatMul
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType, W extends TType> QuantizedMatMul<V> create(Scope scope, Operand<T> a, Operand<U> b, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, DataType<V> Toutput, DataType<W> Tactivation, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType, W extends TType> QuantizedMatMul<V> create(Scope scope, Operand<T> a, Operand<U> b, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Class<V> Toutput, Class<W> Tactivation, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedMatMul", scope.makeOpName("QuantizedMatMul"));
     opBuilder.addInput(a.asOutput());
     opBuilder.addInput(b.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBias.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBias.java
index 52ba1e9576b..4f083e85932 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBias.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBias.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -97,7 +96,7 @@ private Options() {
    * @return a new instance of QuantizedMatMulWithBias
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBias<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, DataType<W> Toutput, Options... options) {
+  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBias<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Class<W> Toutput, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedMatMulWithBias", scope.makeOpName("QuantizedMatMulWithBias"));
     opBuilder.addInput(a.asOutput());
     opBuilder.addInput(b.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndRelu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndRelu.java
index 4d7ba46f49e..c5769d336d9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndRelu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndRelu.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -98,7 +97,7 @@ private Options() {
    * @return a new instance of QuantizedMatMulWithBiasAndRelu
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedMatMulWithBiasAndRelu<V> create(Scope scope, Operand<T> a, Operand<U> b, Operand<TFloat32> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, DataType<V> Toutput, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedMatMulWithBiasAndRelu<V> create(Scope scope, Operand<T> a, Operand<U> b, Operand<TFloat32> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Class<V> Toutput, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedMatMulWithBiasAndRelu", scope.makeOpName("QuantizedMatMulWithBiasAndRelu"));
     opBuilder.addInput(a.asOutput());
     opBuilder.addInput(b.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndReluAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndReluAndRequantize.java
index e5639594805..a580c953e74 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndReluAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/QuantizedMatMulWithBiasAndReluAndRequantize.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -101,7 +100,7 @@ private Options() {
    * @return a new instance of QuantizedMatMulWithBiasAndReluAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBiasAndReluAndRequantize<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<W> Toutput, Options... options) {
+  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBiasAndReluAndRequantize<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<W> Toutput, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedMatMulWithBiasAndReluAndRequantize", scope.makeOpName("QuantizedMatMulWithBiasAndReluAndRequantize"));
     opBuilder.addInput(a.asOutput());
     opBuilder.addInput(b.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixComponents.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixComponents.java
index 332382cec08..3d598e9f450 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixComponents.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixComponents.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,7 +48,7 @@ public final class CSRSparseMatrixComponents<T extends TType> extends RawOp {
    * @return a new instance of CSRSparseMatrixComponents
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> CSRSparseMatrixComponents<T> create(Scope scope, Operand<?> csrSparseMatrix, Operand<TInt32> index, DataType<T> type) {
+  public static <T extends TType> CSRSparseMatrixComponents<T> create(Scope scope, Operand<?> csrSparseMatrix, Operand<TInt32> index, Class<T> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("CSRSparseMatrixComponents", scope.makeOpName("CSRSparseMatrixComponents"));
     opBuilder.addInput(csrSparseMatrix.asOutput());
     opBuilder.addInput(index.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToDense.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToDense.java
index 0d8ceba40d3..6b5fe3c0e82 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToDense.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToDense.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -44,7 +43,7 @@ public final class CSRSparseMatrixToDense<T extends TType> extends RawOp impleme
    * @return a new instance of CSRSparseMatrixToDense
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> CSRSparseMatrixToDense<T> create(Scope scope, Operand<?> sparseInput, DataType<T> type) {
+  public static <T extends TType> CSRSparseMatrixToDense<T> create(Scope scope, Operand<?> sparseInput, Class<T> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("CSRSparseMatrixToDense", scope.makeOpName("CSRSparseMatrixToDense"));
     opBuilder.addInput(sparseInput.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToSparseTensor.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToSparseTensor.java
index 0b5a2d41df8..3e203cac761 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToSparseTensor.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/CSRSparseMatrixToSparseTensor.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,7 +44,7 @@ public final class CSRSparseMatrixToSparseTensor<T extends TType> extends RawOp
    * @return a new instance of CSRSparseMatrixToSparseTensor
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> CSRSparseMatrixToSparseTensor<T> create(Scope scope, Operand<?> sparseMatrix, DataType<T> type) {
+  public static <T extends TType> CSRSparseMatrixToSparseTensor<T> create(Scope scope, Operand<?> sparseMatrix, Class<T> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("CSRSparseMatrixToSparseTensor", scope.makeOpName("CSRSparseMatrixToSparseTensor"));
     opBuilder.addInput(sparseMatrix.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmax.java
index 01cec0f84da..e8043662d36 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmax.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,7 +48,7 @@ public final class SparseMatrixSoftmax extends RawOp implements Operand<TType> {
    * @return a new instance of SparseMatrixSoftmax
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> SparseMatrixSoftmax create(Scope scope, Operand<?> logits, DataType<T> type) {
+  public static <T extends TNumber> SparseMatrixSoftmax create(Scope scope, Operand<?> logits, Class<T> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseMatrixSoftmax", scope.makeOpName("SparseMatrixSoftmax"));
     opBuilder.addInput(logits.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmaxGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmaxGrad.java
index 8ac96d72177..cea37dd4b25 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmaxGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSoftmaxGrad.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -44,7 +43,7 @@ public final class SparseMatrixSoftmaxGrad extends RawOp implements Operand<TTyp
    * @return a new instance of SparseMatrixSoftmaxGrad
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> SparseMatrixSoftmaxGrad create(Scope scope, Operand<?> softmax, Operand<?> gradSoftmax, DataType<T> type) {
+  public static <T extends TNumber> SparseMatrixSoftmaxGrad create(Scope scope, Operand<?> softmax, Operand<?> gradSoftmax, Class<T> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseMatrixSoftmaxGrad", scope.makeOpName("SparseMatrixSoftmaxGrad"));
     opBuilder.addInput(softmax.asOutput());
     opBuilder.addInput(gradSoftmax.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseCholesky.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseCholesky.java
index 82f97e209d9..e70ba2b702b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseCholesky.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseCholesky.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -115,7 +114,7 @@ public final class SparseMatrixSparseCholesky extends RawOp implements Operand<T
    * @return a new instance of SparseMatrixSparseCholesky
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> SparseMatrixSparseCholesky create(Scope scope, Operand<?> input, Operand<TInt32> permutation, DataType<T> type) {
+  public static <T extends TType> SparseMatrixSparseCholesky create(Scope scope, Operand<?> input, Operand<TInt32> permutation, Class<T> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseMatrixSparseCholesky", scope.makeOpName("SparseMatrixSparseCholesky"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(permutation.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseMatMul.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseMatMul.java
index 373ab3a01e2..26ae566f8be 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseMatMul.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixSparseMatMul.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -163,7 +162,7 @@ private Options() {
    * @return a new instance of SparseMatrixSparseMatMul
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> SparseMatrixSparseMatMul create(Scope scope, Operand<?> a, Operand<?> b, DataType<T> type, Options... options) {
+  public static <T extends TType> SparseMatrixSparseMatMul create(Scope scope, Operand<?> a, Operand<?> b, Class<T> type, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseMatrixSparseMatMul", scope.makeOpName("SparseMatrixSparseMatMul"));
     opBuilder.addInput(a.asOutput());
     opBuilder.addInput(b.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixTranspose.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixTranspose.java
index 90e5ba4294f..05e54b0e0e6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixTranspose.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixTranspose.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -65,7 +64,7 @@ private Options() {
    * @return a new instance of SparseMatrixTranspose
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> SparseMatrixTranspose create(Scope scope, Operand<?> input, DataType<T> type, Options... options) {
+  public static <T extends TType> SparseMatrixTranspose create(Scope scope, Operand<?> input, Class<T> type, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseMatrixTranspose", scope.makeOpName("SparseMatrixTranspose"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixZeros.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixZeros.java
index 3c5bef9daed..439c4ce69b7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixZeros.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/linalg/sparse/SparseMatrixZeros.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.linalg.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -43,7 +42,7 @@ public final class SparseMatrixZeros extends RawOp implements Operand<TType> {
    * @return a new instance of SparseMatrixZeros
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> SparseMatrixZeros create(Scope scope, Operand<TInt64> denseShape, DataType<T> type) {
+  public static <T extends TType> SparseMatrixZeros create(Scope scope, Operand<TInt64> denseShape, Class<T> type) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseMatrixZeros", scope.makeOpName("SparseMatrixZeros"));
     opBuilder.addInput(denseShape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Abs.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Abs.java
index 7aa760a535b..7fdc1717833 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Abs.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Abs.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the absolute value of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Angle.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Angle.java
index ce996785568..097a29c35ff 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Angle.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Angle.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -63,7 +62,7 @@ public final class Angle<U extends TNumber> extends RawOp implements Operand<U>
    * @return a new instance of Angle
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Angle<U> create(Scope scope, Operand<T> input, DataType<U> Tout) {
+  public static <U extends TNumber, T extends TType> Angle<U> create(Scope scope, Operand<T> input, Class<U> Tout) {
     OperationBuilder opBuilder = scope.env().opBuilder("Angle", scope.makeOpName("Angle"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -80,7 +79,7 @@ public static <U extends TNumber, T extends TType> Angle<U> create(Scope scope,
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Angle<TFloat32> create(Scope scope, Operand<T> input) {
-    return create(scope, input, TFloat32.DTYPE);
+    return create(scope, input, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMax.java
index b45c1268cb2..42eb6778d24 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMax.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -63,7 +62,7 @@ public final class ArgMax<V extends TNumber> extends RawOp implements Operand<V>
    * @return a new instance of ArgMax
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TType, U extends TNumber> ArgMax<V> create(Scope scope, Operand<T> input, Operand<U> dimension, DataType<V> outputType) {
+  public static <V extends TNumber, T extends TType, U extends TNumber> ArgMax<V> create(Scope scope, Operand<T> input, Operand<U> dimension, Class<V> outputType) {
     OperationBuilder opBuilder = scope.env().opBuilder("ArgMax", scope.makeOpName("ArgMax"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(dimension.asOutput());
@@ -84,7 +83,7 @@ public static <V extends TNumber, T extends TType, U extends TNumber> ArgMax<V>
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType, U extends TNumber> ArgMax<TInt64> create(Scope scope, Operand<T> input, Operand<U> dimension) {
-    return create(scope, input, dimension, TInt64.DTYPE);
+    return create(scope, input, dimension, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMin.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMin.java
index 5a49adecd22..bd6b69d52e1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMin.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ArgMin.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -63,7 +62,7 @@ public final class ArgMin<V extends TNumber> extends RawOp implements Operand<V>
    * @return a new instance of ArgMin
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TType, U extends TNumber> ArgMin<V> create(Scope scope, Operand<T> input, Operand<U> dimension, DataType<V> outputType) {
+  public static <V extends TNumber, T extends TType, U extends TNumber> ArgMin<V> create(Scope scope, Operand<T> input, Operand<U> dimension, Class<V> outputType) {
     OperationBuilder opBuilder = scope.env().opBuilder("ArgMin", scope.makeOpName("ArgMin"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(dimension.asOutput());
@@ -84,7 +83,7 @@ public static <V extends TNumber, T extends TType, U extends TNumber> ArgMin<V>
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType, U extends TNumber> ArgMin<TInt64> create(Scope scope, Operand<T> input, Operand<U> dimension) {
-    return create(scope, input, dimension, TInt64.DTYPE);
+    return create(scope, input, dimension, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Atan2.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Atan2.java
index d49bc324cb3..a70043da3a3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Atan2.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Atan2.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes arctangent of `y/x` element-wise, respecting signs of the arguments.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0.java
index 45dcd2b8e4c..ecf7aab945c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0e.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0e.java
index 816933ceb12..d274265faf0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0e.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI0e.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1.java
index 148758aa5a4..b836404991e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1e.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1e.java
index 3529c4a0bed..f8d40882585 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1e.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/BesselI1e.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Betainc.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Betainc.java
index 53ecd2ed396..e28da25745f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Betainc.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Betainc.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Compute the regularized incomplete beta integral \\(I_x(a, b)\\).
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Bincount.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Bincount.java
index 4217584c4fb..565a006c3b4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Bincount.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Bincount.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Counts the number of occurrences of each value in an integer array.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ceil.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ceil.java
index c97e5b5ae5d..bda7247d508 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ceil.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ceil.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns element-wise smallest integer not less than x.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ComplexAbs.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ComplexAbs.java
index 317744e519a..b899c47845a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ComplexAbs.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/ComplexAbs.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -52,7 +51,7 @@ public final class ComplexAbs<U extends TNumber> extends RawOp implements Operan
    * @return a new instance of ComplexAbs
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> ComplexAbs<U> create(Scope scope, Operand<T> x, DataType<U> Tout) {
+  public static <U extends TNumber, T extends TType> ComplexAbs<U> create(Scope scope, Operand<T> x, Class<U> Tout) {
     OperationBuilder opBuilder = scope.env().opBuilder("ComplexAbs", scope.makeOpName("ComplexAbs"));
     opBuilder.addInput(x.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -69,7 +68,7 @@ public static <U extends TNumber, T extends TType> ComplexAbs<U> create(Scope sc
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> ComplexAbs<TFloat32> create(Scope scope, Operand<T> x) {
-    return create(scope, x, TFloat32.DTYPE);
+    return create(scope, x, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/CumulativeLogsumexp.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/CumulativeLogsumexp.java
index 3c8e6f89870..4f199369473 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/CumulativeLogsumexp.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/CumulativeLogsumexp.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Compute the cumulative product of the tensor `x` along `axis`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/DenseBincount.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/DenseBincount.java
index e38d559f4ae..edbc39b49cb 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/DenseBincount.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/DenseBincount.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Counts the number of occurrences of each value in an integer array.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Digamma.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Digamma.java
index 2fd494f637c..571ade7795b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Digamma.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Digamma.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes Psi, the derivative of Lgamma (the log of the absolute value of
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erf.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erf.java
index 9f86126bff4..ff1c2d343e8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erf.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erf.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the Gauss error function of `x` element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erfc.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erfc.java
index b94fecf9ede..31826d91508 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erfc.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Erfc.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the complementary error function of `x` element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Floor.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Floor.java
index ac8d7a8ffba..acd6f3a1d2b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Floor.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Floor.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns element-wise largest integer not greater than x.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/FloorMod.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/FloorMod.java
index 715628b57a3..ae4e933ccea 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/FloorMod.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/FloorMod.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns element-wise remainder of division. When `x < 0` xor `y < 0` is
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Greater.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Greater.java
index e8ef3811fd1..a0f100e6b03 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Greater.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Greater.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the truth value of (x > y) element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/GreaterEqual.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/GreaterEqual.java
index 11a83743031..fa67314267f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/GreaterEqual.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/GreaterEqual.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the truth value of (x >= y) element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igamma.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igamma.java
index a4687ef3ab1..6d757ea96c6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igamma.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igamma.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Compute the lower regularized incomplete Gamma function `P(a, x)`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IgammaGradA.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IgammaGradA.java
index e57b8d886e1..9e757f8e01d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IgammaGradA.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IgammaGradA.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of `igamma(a, x)` wrt `a`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igammac.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igammac.java
index 1e2cb0cf0f7..dec4671fe76 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igammac.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Igammac.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Compute the upper regularized incomplete Gamma function `Q(a, x)`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Imag.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Imag.java
index 48a8a7107a3..9c4ce01655e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Imag.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Imag.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -59,7 +58,7 @@ public final class Imag<U extends TNumber> extends RawOp implements Operand<U> {
    * @return a new instance of Imag
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Imag<U> create(Scope scope, Operand<T> input, DataType<U> Tout) {
+  public static <U extends TNumber, T extends TType> Imag<U> create(Scope scope, Operand<T> input, Class<U> Tout) {
     OperationBuilder opBuilder = scope.env().opBuilder("Imag", scope.makeOpName("Imag"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -76,7 +75,7 @@ public static <U extends TNumber, T extends TType> Imag<U> create(Scope scope, O
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Imag<TFloat32> create(Scope scope, Operand<T> input) {
-    return create(scope, input, TFloat32.DTYPE);
+    return create(scope, input, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/InvertPermutation.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/InvertPermutation.java
index de9e9c46932..859d648b2c5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/InvertPermutation.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/InvertPermutation.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the inverse permutation of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsFinite.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsFinite.java
index ffe9bb99f10..b6536046d53 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsFinite.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsFinite.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns which elements of x are finite.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsInf.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsInf.java
index 3d36e9fbfe9..cde3b798d16 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsInf.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsInf.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns which elements of x are Inf.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsNan.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsNan.java
index b58205a005c..0ecb4836cee 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsNan.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/IsNan.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns which elements of x are NaN.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Less.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Less.java
index 3d796296f01..cf696b5a759 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Less.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Less.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the truth value of (x < y) element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/LessEqual.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/LessEqual.java
index 618cd7a9866..4b4b85a8198 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/LessEqual.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/LessEqual.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the truth value of (x <= y) element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Lgamma.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Lgamma.java
index 9bc18b1b3b8..8b3070659e1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Lgamma.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Lgamma.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the log of the absolute value of `Gamma(x)` element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Maximum.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Maximum.java
index 9f72150c5d3..b6360e35d91 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Maximum.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Maximum.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the max of x and y (i.e. x > y ? x : y) element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Minimum.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Minimum.java
index e11b6e484fc..7a3e003bfad 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Minimum.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Minimum.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the min of x and y (i.e. x < y ? x : y) element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Mod.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Mod.java
index 7ba98b81b39..0502e287fc5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Mod.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Mod.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns element-wise remainder of division. This emulates C semantics in that
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ndtri.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ndtri.java
index 55dcf0c434d..eec0a0d1c9e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ndtri.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Ndtri.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/NextAfter.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/NextAfter.java
index 8fa53306eed..7f494a874d5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/NextAfter.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/NextAfter.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the next representable value of `x1` in the direction of `x2`, element-wise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Polygamma.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Polygamma.java
index d0021efc2ac..173ce8cd757 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Polygamma.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Polygamma.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Compute the polygamma function \\(\psi^{(n)}(x)\\).
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/PopulationCount.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/PopulationCount.java
index bc721897426..68c9ebf055b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/PopulationCount.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/PopulationCount.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TUint8;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes element-wise population count (a.k.a. popcount, bitsum, bitcount).
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedAdd.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedAdd.java
index ccd6d2e5f98..2c956969e8d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedAdd.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedAdd.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,7 +50,7 @@ public final class QuantizedAdd<V extends TType> extends RawOp {
    * @return a new instance of QuantizedAdd
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedAdd<V> create(Scope scope, Operand<T> x, Operand<U> y, Operand<TFloat32> minX, Operand<TFloat32> maxX, Operand<TFloat32> minY, Operand<TFloat32> maxY, DataType<V> Toutput) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedAdd<V> create(Scope scope, Operand<T> x, Operand<U> y, Operand<TFloat32> minX, Operand<TFloat32> maxX, Operand<TFloat32> minY, Operand<TFloat32> maxY, Class<V> Toutput) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedAdd", scope.makeOpName("QuantizedAdd"));
     opBuilder.addInput(x.asOutput());
     opBuilder.addInput(y.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedMul.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedMul.java
index e16dc423a63..bd666426bd1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedMul.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/QuantizedMul.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,7 +50,7 @@ public final class QuantizedMul<V extends TType> extends RawOp {
    * @return a new instance of QuantizedMul
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedMul<V> create(Scope scope, Operand<T> x, Operand<U> y, Operand<TFloat32> minX, Operand<TFloat32> maxX, Operand<TFloat32> minY, Operand<TFloat32> maxY, DataType<V> Toutput) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedMul<V> create(Scope scope, Operand<T> x, Operand<U> y, Operand<TFloat32> minX, Operand<TFloat32> maxX, Operand<TFloat32> minY, Operand<TFloat32> maxY, Class<V> Toutput) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedMul", scope.makeOpName("QuantizedMul"));
     opBuilder.addInput(x.asOutput());
     opBuilder.addInput(y.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Real.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Real.java
index e113597afd8..bd00e9a4ae6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Real.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Real.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -59,7 +58,7 @@ public final class Real<U extends TNumber> extends RawOp implements Operand<U> {
    * @return a new instance of Real
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Real<U> create(Scope scope, Operand<T> input, DataType<U> Tout) {
+  public static <U extends TNumber, T extends TType> Real<U> create(Scope scope, Operand<T> input, Class<U> Tout) {
     OperationBuilder opBuilder = scope.env().opBuilder("Real", scope.makeOpName("Real"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -76,7 +75,7 @@ public static <U extends TNumber, T extends TType> Real<U> create(Scope scope, O
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Real<TFloat32> create(Scope scope, Operand<T> input) {
-    return create(scope, input, TFloat32.DTYPE);
+    return create(scope, input, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/RequantizePerChannel.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/RequantizePerChannel.java
index 53a6201151e..515da442497 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/RequantizePerChannel.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/RequantizePerChannel.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,7 +48,7 @@ public final class RequantizePerChannel<U extends TType> extends RawOp {
    * @return a new instance of RequantizePerChannel
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> RequantizePerChannel<U> create(Scope scope, Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, Operand<TFloat32> requestedOutputMin, Operand<TFloat32> requestedOutputMax, DataType<U> outType) {
+  public static <U extends TType, T extends TType> RequantizePerChannel<U> create(Scope scope, Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, Operand<TFloat32> requestedOutputMin, Operand<TFloat32> requestedOutputMax, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("RequantizePerChannel", scope.makeOpName("RequantizePerChannel"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(inputMin.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Rint.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Rint.java
index b7b152248a8..849508eba02 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Rint.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Rint.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns element-wise integer closest to x.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMax.java
index fdd40054420..e044212668b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMax.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the maximum along segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMin.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMin.java
index b7ab590e976..b628c6f5eb1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMin.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SegmentMin.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the minimum along segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SobolSample.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SobolSample.java
index 131cb7ed792..e7c9c43d20d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SobolSample.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SobolSample.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.math;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Generates points from the Sobol sequence.
@@ -54,7 +52,7 @@ public final class SobolSample<T extends TNumber> extends RawOp implements Opera
    * @return a new instance of SobolSample
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> SobolSample<T> create(Scope scope, Operand<TInt32> dim, Operand<TInt32> numResults, Operand<TInt32> skip, DataType<T> dtype) {
+  public static <T extends TNumber> SobolSample<T> create(Scope scope, Operand<TInt32> dim, Operand<TInt32> numResults, Operand<TInt32> skip, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("SobolSample", scope.makeOpName("SobolSample"));
     opBuilder.addInput(dim.asOutput());
     opBuilder.addInput(numResults.asOutput());
@@ -77,7 +75,7 @@ public static <T extends TNumber> SobolSample<T> create(Scope scope, Operand<TIn
    */
   @Endpoint(describeByClass = true)
   public static SobolSample<TFloat32> create(Scope scope, Operand<TInt32> dim, Operand<TInt32> numResults, Operand<TInt32> skip) {
-    return create(scope, dim, numResults, skip, TFloat32.DTYPE);
+    return create(scope, dim, numResults, skip, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Softplus.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Softplus.java
index 79adcf30d76..5065f948cc1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Softplus.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Softplus.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes softplus: `log(exp(features) + 1)`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SoftplusGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SoftplusGrad.java
index 90174757bc3..100243c60e4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SoftplusGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/SoftplusGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes softplus gradients for a softplus operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/TruncateMod.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/TruncateMod.java
index 48c92574eb4..0fb4397c881 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/TruncateMod.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/TruncateMod.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns element-wise remainder of division. This emulates C semantics in that
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMax.java
index 582113e3320..733e3105846 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMax.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the maximum along segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMin.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMin.java
index 397ac5706a0..408995715a0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMin.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/UnsortedSegmentMin.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the minimum along segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Zeta.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Zeta.java
index 0f3cee188fc..576af0a1e6e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Zeta.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/Zeta.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Compute the Hurwitz zeta function \\(\zeta(x, q)\\).
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/erfinv.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/erfinv.java
index de2a4482f65..e74d429e129 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/erfinv.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/erfinv.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ0.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ0.java
index 8d2184a49cb..c49c2d69350 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ0.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ0.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ1.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ1.java
index d8f9621a36c..fb7f29046de 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ1.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselJ1.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0.java
index eaae243f83f..945b617b1dc 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0e.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0e.java
index c57ae64e233..d1a014d4959 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0e.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK0e.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1.java
index 1858d25fe3d..0cb60fa3d88 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1e.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1e.java
index e4a5cc23efd..b01a57da175 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1e.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselK1e.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY0.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY0.java
index 9228d1b6145..619147cbb6b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY0.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY0.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY1.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY1.java
index 0461416b808..51eb6e377f2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY1.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/BesselY1.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Dawsn.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Dawsn.java
index 74388434149..529514c9bf1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Dawsn.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Dawsn.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Expint.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Expint.java
index b36c55fdeb6..099cd544ada 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Expint.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Expint.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelCos.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelCos.java
index bb9a9f47e78..5d5ce736e42 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelCos.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelCos.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelSin.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelSin.java
index 36681c87678..5914fe4da76 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelSin.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/FresnelSin.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Spence.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Spence.java
index ed613a28b1a..faf21f53510 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Spence.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/math/special/Spence.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <T> data type for {@code y()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool.java
index 527d1a49713..a27f1aa70c7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs average pooling on the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3d.java
index 87467c8a982..177227c0a52 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3d.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs 3D average pooling on the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3dGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3dGrad.java
index 3e34c87f9b6..c23a0c5360c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3dGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPool3dGrad.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients of average pooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPoolGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPoolGrad.java
index 7ec252eaac0..ec149807546 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPoolGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/AvgPoolGrad.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients of the average pooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTM.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTM.java
index 79de4f2f88c..c40f2eeea4b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTM.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTM.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the LSTM cell forward propagation for all the time steps.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTMGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTMGrad.java
index 4e49f23cb6e..f7a17b1bf76 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTMGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/BlockLSTMGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the LSTM cell backward propagation for the entire time sequence.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2d.java
index 79f2022d807..70a72bc78da 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2d.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes a 2-D convolution given 4-D `input` and `filter` tensors.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropFilter.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropFilter.java
index 9c145dc3d60..709d59dee28 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropFilter.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropFilter.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradients of convolution with respect to the filter.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropInput.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropInput.java
index 492c4f2aebb..b725823e8db 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropInput.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv2dBackpropInput.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradients of convolution with respect to the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3d.java
index 94ae4ffd2b5..0720ef0dce5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3d.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes a 3-D convolution given 5-D `input` and `filter` tensors.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropFilter.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropFilter.java
index 0d73e491717..1de81e9f781 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropFilter.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropFilter.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradients of 3-D convolution with respect to the filter.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropInput.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropInput.java
index 8b153890811..1f71f6b64f3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropInput.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Conv3dBackpropInput.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradients of 3-D convolution with respect to the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcBeamSearchDecoder.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcBeamSearchDecoder.java
index 96f179641e7..fadc3985f38 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcBeamSearchDecoder.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcBeamSearchDecoder.java
@@ -30,7 +30,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs beam search decoding on the logits given in input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcGreedyDecoder.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcGreedyDecoder.java
index b8b35b8ceaa..2c985fb4e03 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcGreedyDecoder.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcGreedyDecoder.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs greedy decoding on the logits given in inputs.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcLoss.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcLoss.java
index ceb60a4baf7..44acdfbad19 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcLoss.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CtcLoss.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Calculates the CTC Loss (log probability) for each batch entry.  Also calculates
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNN.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNN.java
index 7c179d7e578..44495c8b7a7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNN.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNN.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * A RNN backed by cuDNN.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNBackprop.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNBackprop.java
index 3719c186716..9e51b12fee7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNBackprop.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNBackprop.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Backprop step of CudnnRNNV3.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNCanonicalToParams.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNCanonicalToParams.java
index 155cfd2c0a4..38df41c251a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNCanonicalToParams.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNCanonicalToParams.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Converts CudnnRNN params from canonical form to usable form. It supports the projection in LSTM.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNParamsToCanonical.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNParamsToCanonical.java
index ea575f2d7a1..a6fcda3786e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNParamsToCanonical.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRNNParamsToCanonical.java
@@ -29,7 +29,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Retrieves CudnnRNN params in canonical form. It supports the projection in LSTM.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRnnParamsSize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRnnParamsSize.java
index 94421d0aa6e..dd3f4ba6392 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRnnParamsSize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/CudnnRnnParamsSize.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.nn;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes size of weights that can be used by a Cudnn RNN model.
@@ -146,7 +144,7 @@ private Options() {
    * @return a new instance of CudnnRnnParamsSize
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TNumber> CudnnRnnParamsSize<U> create(Scope scope, Operand<TInt32> numLayers, Operand<TInt32> numUnits, Operand<TInt32> inputSize, DataType<T> T, DataType<U> S, Options... options) {
+  public static <U extends TNumber, T extends TNumber> CudnnRnnParamsSize<U> create(Scope scope, Operand<TInt32> numLayers, Operand<TInt32> numUnits, Operand<TInt32> inputSize, Class<T> T, Class<U> S, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("CudnnRNNParamsSize", scope.makeOpName("CudnnRnnParamsSize"));
     opBuilder.addInput(numLayers.asOutput());
     opBuilder.addInput(numUnits.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatDimMap.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatDimMap.java
index 6b1ff40761b..3eb8a46d310 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatDimMap.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatDimMap.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the dimension index in the destination data format given the one in
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatVecPermute.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatVecPermute.java
index ba218c1923f..0ba1ddeb4ba 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatVecPermute.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DataFormatVecPermute.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the permuted vector/tensor in the destination data format given the
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNative.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNative.java
index 6549d8b9dfc..2aff519c733 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNative.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNative.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes a 2-D depthwise convolution given 4-D `input` and `filter` tensors.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropFilter.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropFilter.java
index fadfdacc823..ac51c06e737 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropFilter.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropFilter.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradients of depthwise convolution with respect to the filter.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropInput.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropInput.java
index d2e3e733a01..26db9841d7d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropInput.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/DepthwiseConv2dNativeBackpropInput.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradients of depthwise convolution with respect to the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2d.java
index c7135b20361..e6bd8cc5b6a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2d.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the grayscale dilation of 4-D `input` and 3-D `filter` tensors.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropFilter.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropFilter.java
index 9254ff8c285..04705f53024 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropFilter.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropFilter.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of morphological 2-D dilation with respect to the filter.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropInput.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropInput.java
index 525e06182c5..1636545164c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropInput.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Dilation2dBackpropInput.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the gradient of morphological 2-D dilation with respect to the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Elu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Elu.java
index daeef97895b..faa70f04b98 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Elu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Elu.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes exponential linear: `exp(features) - 1` if < 0, `features` otherwise.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/EluGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/EluGrad.java
index 664475879a0..9ead32623ab 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/EluGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/EluGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients for the exponential linear (Elu) operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPool.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPool.java
index 101ea21ec8c..93a477ef25e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPool.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPool.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs fractional average pooling on the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPoolGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPoolGrad.java
index 03ae0136311..7a4258d18c8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPoolGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalAvgPoolGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradient of the FractionalAvgPool function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPool.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPool.java
index a621e037740..9ca640c804b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPool.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPool.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs fractional max pooling on the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPoolGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPoolGrad.java
index 1b2bcc62dbf..952c8ce286a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPoolGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FractionalMaxPoolGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradient of the FractionalMaxPool function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNorm.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNorm.java
index 04e8d71e1e5..0d2b3a6dfe9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNorm.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNorm.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Batch normalization.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNormGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNormGrad.java
index 4d2ecc74a4f..8e9f616adcb 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNormGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedBatchNormGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Gradient for batch normalization.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedPadConv2d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedPadConv2d.java
index 2d3de921f0b..89713030f7b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedPadConv2d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedPadConv2d.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs a padding as a preprocess during a convolution.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedResizeAndPadConv2d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedResizeAndPadConv2d.java
index 02f82242c7a..b89af628392 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedResizeAndPadConv2d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/FusedResizeAndPadConv2d.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs a resize and padding as a preprocess during a convolution.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCell.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCell.java
index 446f43cfb0c..fad66727252 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCell.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCell.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the GRU cell forward propagation for 1 time step.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCellGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCellGrad.java
index 4a6b72b8e4c..69c4295d5c4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCellGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/GRUBlockCellGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the GRU cell back-propagation for 1 time step.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/InTopK.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/InTopK.java
index d38388a3fd5..ea2a2652b83 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/InTopK.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/InTopK.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Says whether the targets are in the top `K` predictions.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/L2Loss.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/L2Loss.java
index b38d9e99d96..eb38a9baa65 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/L2Loss.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/L2Loss.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * L2 Loss.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCell.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCell.java
index ab5d1b8aee2..160d2bc89be 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCell.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCell.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the LSTM cell forward propagation for 1 time step.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCellGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCellGrad.java
index 3635ec433d6..3a36077d6ca 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCellGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LSTMBlockCellGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the LSTM cell backward propagation for 1 timestep.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LeakyRelu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LeakyRelu.java
index 8ca3f540cad..96f44eacb5a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LeakyRelu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LeakyRelu.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes rectified linear: `max(features, features * alpha)`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalization.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalization.java
index a560c9cfb58..e7d72196b4a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalization.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalization.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Local Response Normalization.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalizationGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalizationGrad.java
index 08f10c12f26..d347bec7237 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalizationGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LocalResponseNormalizationGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Gradients for Local Response Normalization.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LogSoftmax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LogSoftmax.java
index bd1bff467a1..e9e123e516d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LogSoftmax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/LogSoftmax.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes log softmax activations.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3d.java
index 61bf43372cc..320f38a084e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3d.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs 3D max pooling on the input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGrad.java
index 2a4fe5e61f9..7f2938e8827 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients of 3D max pooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGradGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGradGrad.java
index 4243e1e5143..ef4c546c41e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGradGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPool3dGradGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes second-order gradients of the maxpooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGrad.java
index 6f5cbd2ce64..e391585b657 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients of the maxpooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGrad.java
index 4f9255e9286..2653f1d51c2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes second-order gradients of the maxpooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGradWithArgmax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGradWithArgmax.java
index 6258024daa9..a15222e7c4f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGradWithArgmax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradGradWithArgmax.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes second-order gradients of the maxpooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradWithArgmax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradWithArgmax.java
index 329c88b4d33..c51275f68f2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradWithArgmax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolGradWithArgmax.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients of the maxpooling function.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolWithArgmax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolWithArgmax.java
index 63f92de1fea..841e598ee3a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolWithArgmax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/MaxPoolWithArgmax.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs max pooling on the input and outputs both max values and indices.
@@ -83,7 +81,7 @@ private Options() {
    * @return a new instance of MaxPoolWithArgmax
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber, U extends TNumber> MaxPoolWithArgmax<T, U> create(Scope scope, Operand<T> input, List<Long> ksize, List<Long> strides, DataType<U> Targmax, String padding, Options... options) {
+  public static <T extends TNumber, U extends TNumber> MaxPoolWithArgmax<T, U> create(Scope scope, Operand<T> input, List<Long> ksize, List<Long> strides, Class<U> Targmax, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("MaxPoolWithArgmax", scope.makeOpName("MaxPoolWithArgmax"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -123,7 +121,7 @@ public static <T extends TNumber, U extends TNumber> MaxPoolWithArgmax<T, U> cre
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber> MaxPoolWithArgmax<T, TInt64> create(Scope scope, Operand<T> input, List<Long> ksize, List<Long> strides, String padding, Options... options) {
-    return create(scope, input, ksize, strides, TInt64.DTYPE, padding, options);
+    return create(scope, input, ksize, strides, TInt64.class, padding, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/NthElement.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/NthElement.java
index 340cd4b98f7..35f827f5d01 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/NthElement.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/NthElement.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Finds values of the `n`-th order statistic for the last dimension.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBatchNormWithGlobalNormalization.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBatchNormWithGlobalNormalization.java
index 41cc16017fc..57b2af3a66d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBatchNormWithGlobalNormalization.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBatchNormWithGlobalNormalization.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.nn;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -73,7 +72,7 @@ public final class QuantizedBatchNormWithGlobalNormalization<U extends TType> ex
    * @return a new instance of QuantizedBatchNormWithGlobalNormalization
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> QuantizedBatchNormWithGlobalNormalization<U> create(Scope scope, Operand<T> t, Operand<TFloat32> tMin, Operand<TFloat32> tMax, Operand<T> m, Operand<TFloat32> mMin, Operand<TFloat32> mMax, Operand<T> v, Operand<TFloat32> vMin, Operand<TFloat32> vMax, Operand<T> beta, Operand<TFloat32> betaMin, Operand<TFloat32> betaMax, Operand<T> gamma, Operand<TFloat32> gammaMin, Operand<TFloat32> gammaMax, DataType<U> outType, Float varianceEpsilon, Boolean scaleAfterNormalization) {
+  public static <U extends TType, T extends TType> QuantizedBatchNormWithGlobalNormalization<U> create(Scope scope, Operand<T> t, Operand<TFloat32> tMin, Operand<TFloat32> tMax, Operand<T> m, Operand<TFloat32> mMin, Operand<TFloat32> mMax, Operand<T> v, Operand<TFloat32> vMin, Operand<TFloat32> vMax, Operand<T> beta, Operand<TFloat32> betaMin, Operand<TFloat32> betaMax, Operand<T> gamma, Operand<TFloat32> gammaMin, Operand<TFloat32> gammaMax, Class<U> outType, Float varianceEpsilon, Boolean scaleAfterNormalization) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedBatchNormWithGlobalNormalization", scope.makeOpName("QuantizedBatchNormWithGlobalNormalization"));
     opBuilder.addInput(t.asOutput());
     opBuilder.addInput(tMin.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBiasAdd.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBiasAdd.java
index 62a8002ad49..b39db777a05 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBiasAdd.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedBiasAdd.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.nn;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -53,7 +52,7 @@ public final class QuantizedBiasAdd<V extends TType> extends RawOp {
    * @return a new instance of QuantizedBiasAdd
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedBiasAdd<V> create(Scope scope, Operand<T> input, Operand<U> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minBias, Operand<TFloat32> maxBias, DataType<V> outType) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedBiasAdd<V> create(Scope scope, Operand<T> input, Operand<U> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minBias, Operand<TFloat32> maxBias, Class<V> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedBiasAdd", scope.makeOpName("QuantizedBiasAdd"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(bias.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRelu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRelu.java
index 961a4245155..362a64d595e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRelu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRelu.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -80,7 +79,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DAndRelu
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DAndRelu", scope.makeOpName("QuantizedConv2DAndRelu"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndReluAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndReluAndRequantize.java
index 8b231edc09b..f55cae7b48f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndReluAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndReluAndRequantize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -82,7 +81,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DAndReluAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DAndReluAndRequantize<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DAndReluAndRequantize<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DAndReluAndRequantize", scope.makeOpName("QuantizedConv2DAndReluAndRequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRequantize.java
index 8c9f1247d97..30f303ae8f9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DAndRequantize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -82,7 +81,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DAndRequantize<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DAndRequantize<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DAndRequantize", scope.makeOpName("QuantizedConv2DAndRequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DPerChannel.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DPerChannel.java
index efc0132e2a8..23019817b6d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DPerChannel.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DPerChannel.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -73,7 +72,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DPerChannel
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DPerChannel<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DPerChannel<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DPerChannel", scope.makeOpName("QuantizedConv2DPerChannel"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBias.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBias.java
index bc78b5b4ec1..a0e9fab70ab 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBias.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBias.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -81,7 +80,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DWithBias
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DWithBias<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DWithBias<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DWithBias", scope.makeOpName("QuantizedConv2DWithBias"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRelu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRelu.java
index 719150ee139..b0288a5adf0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRelu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRelu.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -81,7 +80,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DWithBiasAndRelu
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DWithBiasAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DWithBiasAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DWithBiasAndRelu", scope.makeOpName("QuantizedConv2DWithBiasAndRelu"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndReluAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndReluAndRequantize.java
index a61cab41d5e..c06f4ebe73c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndReluAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndReluAndRequantize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -83,7 +82,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DWithBiasAndReluAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedConv2DWithBiasAndReluAndRequantize<W> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<W> outType, List<Long> strides, String padding, Options... options) {
+  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedConv2DWithBiasAndReluAndRequantize<W> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<W> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DWithBiasAndReluAndRequantize", scope.makeOpName("QuantizedConv2DWithBiasAndReluAndRequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRequantize.java
index af89fd48962..81d1f4497a0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasAndRequantize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -83,7 +82,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DWithBiasAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedConv2DWithBiasAndRequantize<W> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<W> outType, List<Long> strides, String padding, Options... options) {
+  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedConv2DWithBiasAndRequantize<W> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<W> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DWithBiasAndRequantize", scope.makeOpName("QuantizedConv2DWithBiasAndRequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSignedSumAndReluAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSignedSumAndReluAndRequantize.java
index cfa7770cf25..3c60a3f3a1a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSignedSumAndReluAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSignedSumAndReluAndRequantize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -86,7 +85,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DWithBiasSignedSumAndReluAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <X extends TType, T extends TType, U extends TType, V extends TType, W extends TType> QuantizedConv2DWithBiasSignedSumAndReluAndRequantize<X> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Operand<W> summand, Operand<TFloat32> minSummand, Operand<TFloat32> maxSummand, DataType<X> outType, List<Long> strides, String padding, Options... options) {
+  public static <X extends TType, T extends TType, U extends TType, V extends TType, W extends TType> QuantizedConv2DWithBiasSignedSumAndReluAndRequantize<X> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Operand<W> summand, Operand<TFloat32> minSummand, Operand<TFloat32> maxSummand, Class<X> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DWithBiasSignedSumAndReluAndRequantize", scope.makeOpName("QuantizedConv2DWithBiasSignedSumAndReluAndRequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndRelu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndRelu.java
index 1096fa12578..0788d608ac1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndRelu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndRelu.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -82,7 +81,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DWithBiasSumAndRelu
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DWithBiasSumAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> summand, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2DWithBiasSumAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> summand, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DWithBiasSumAndRelu", scope.makeOpName("QuantizedConv2DWithBiasSumAndRelu"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndReluAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndReluAndRequantize.java
index 55adb0e016d..91c903bad91 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndReluAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2DWithBiasSumAndReluAndRequantize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -86,7 +85,7 @@ private Options() {
    * @return a new instance of QuantizedConv2DWithBiasSumAndReluAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <X extends TType, T extends TType, U extends TType, V extends TType, W extends TType> QuantizedConv2DWithBiasSumAndReluAndRequantize<X> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Operand<W> summand, Operand<TFloat32> minSummand, Operand<TFloat32> maxSummand, DataType<X> outType, List<Long> strides, String padding, Options... options) {
+  public static <X extends TType, T extends TType, U extends TType, V extends TType, W extends TType> QuantizedConv2DWithBiasSumAndReluAndRequantize<X> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Operand<W> summand, Operand<TFloat32> minSummand, Operand<TFloat32> maxSummand, Class<X> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2DWithBiasSumAndReluAndRequantize", scope.makeOpName("QuantizedConv2DWithBiasSumAndReluAndRequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2d.java
index 39913922bdb..ffd66fc3f23 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedConv2d.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -84,7 +83,7 @@ private Options() {
    * @return a new instance of QuantizedConv2d
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2d<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedConv2d<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedConv2D", scope.makeOpName("QuantizedConv2d"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2D.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2D.java
index f64d9a7efbb..1e0a8c38e0f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2D.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2D.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -73,7 +72,7 @@ private Options() {
    * @return a new instance of QuantizedDepthwiseConv2D
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedDepthwiseConv2D<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedDepthwiseConv2D<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedDepthwiseConv2D", scope.makeOpName("QuantizedDepthwiseConv2D"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBias.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBias.java
index be5e2bb8657..b0d29c64e39 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBias.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBias.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -74,7 +73,7 @@ private Options() {
    * @return a new instance of QuantizedDepthwiseConv2DWithBias
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedDepthwiseConv2DWithBias<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedDepthwiseConv2DWithBias<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedDepthwiseConv2DWithBias", scope.makeOpName("QuantizedDepthwiseConv2DWithBias"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndRelu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndRelu.java
index 8abd12b865f..e88bddc4f83 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndRelu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndRelu.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -83,7 +82,7 @@ private Options() {
    * @return a new instance of QuantizedDepthwiseConv2DWithBiasAndRelu
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TType, T extends TType, U extends TType> QuantizedDepthwiseConv2DWithBiasAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, DataType<V> outType, List<Long> strides, String padding, Options... options) {
+  public static <V extends TType, T extends TType, U extends TType> QuantizedDepthwiseConv2DWithBiasAndRelu<V> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<TFloat32> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Class<V> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedDepthwiseConv2DWithBiasAndRelu", scope.makeOpName("QuantizedDepthwiseConv2DWithBiasAndRelu"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize.java
index 78c8048266e..3430d0a6e2f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize.java
@@ -18,7 +18,6 @@
 package org.tensorflow.op.nn;
 
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -85,7 +84,7 @@ private Options() {
    * @return a new instance of QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize<W> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<W> outType, List<Long> strides, String padding, Options... options) {
+  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize<W> create(Scope scope, Operand<T> input, Operand<U> filter, Operand<V> bias, Operand<TFloat32> minInput, Operand<TFloat32> maxInput, Operand<TFloat32> minFilter, Operand<TFloat32> maxFilter, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<W> outType, List<Long> strides, String padding, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize", scope.makeOpName("QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(filter.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu.java
index 308e14e9512..8881d2eb824 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.nn;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,7 +47,7 @@ public final class QuantizedRelu<U extends TType> extends RawOp {
    * @return a new instance of QuantizedRelu
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> QuantizedRelu<U> create(Scope scope, Operand<T> features, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, DataType<U> outType) {
+  public static <U extends TType, T extends TType> QuantizedRelu<U> create(Scope scope, Operand<T> features, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedRelu", scope.makeOpName("QuantizedRelu"));
     opBuilder.addInput(features.asOutput());
     opBuilder.addInput(minFeatures.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu6.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu6.java
index 46f04fd5722..2153c3561df 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu6.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedRelu6.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.nn;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -48,7 +47,7 @@ public final class QuantizedRelu6<U extends TType> extends RawOp {
    * @return a new instance of QuantizedRelu6
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> QuantizedRelu6<U> create(Scope scope, Operand<T> features, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, DataType<U> outType) {
+  public static <U extends TType, T extends TType> QuantizedRelu6<U> create(Scope scope, Operand<T> features, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedRelu6", scope.makeOpName("QuantizedRelu6"));
     opBuilder.addInput(features.asOutput());
     opBuilder.addInput(minFeatures.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedReluX.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedReluX.java
index e47a6a5c043..9db60f2174c 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedReluX.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/QuantizedReluX.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.nn;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,7 +48,7 @@ public final class QuantizedReluX<U extends TType> extends RawOp {
    * @return a new instance of QuantizedReluX
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> QuantizedReluX<U> create(Scope scope, Operand<T> features, Operand<TFloat32> maxValue, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, DataType<U> outType) {
+  public static <U extends TType, T extends TType> QuantizedReluX<U> create(Scope scope, Operand<T> features, Operand<TFloat32> maxValue, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedReluX", scope.makeOpName("QuantizedReluX"));
     opBuilder.addInput(features.asOutput());
     opBuilder.addInput(maxValue.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6.java
index e4e674b7e35..f227e995b2e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes rectified linear 6: `min(max(features, 0), 6)`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6Grad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6Grad.java
index cf4edc9debf..043f3b5a5d7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6Grad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Relu6Grad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes rectified linear 6 gradients for a Relu6 operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/ReluGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/ReluGrad.java
index d1c7cf7d44c..b41caedda64 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/ReluGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/ReluGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes rectified linear gradients for a Relu operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Selu.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Selu.java
index 2ce8484b299..355222c25fa 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Selu.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Selu.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes scaled exponential linear: `scale * alpha * (exp(features) - 1)`
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SeluGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SeluGrad.java
index 514ad00b38d..6bf70f4038d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SeluGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SeluGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients for the scaled exponential linear (Selu) operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softmax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softmax.java
index d9971b6667d..093195aea28 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softmax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softmax.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes softmax activations.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softsign.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softsign.java
index e4276c679a3..0b2e3a639e9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softsign.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/Softsign.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes softsign: `features / (abs(features) + 1)`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SoftsignGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SoftsignGrad.java
index 220763e57a2..31367f06df5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SoftsignGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SoftsignGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes softsign gradients for a softsign operation.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/TopK.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/TopK.java
index acd4ba679f7..a8bfdbea715 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/TopK.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/TopK.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Finds values and indices of the `k` largest elements for the last dimension.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SoftmaxCrossEntropyWithLogits.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SoftmaxCrossEntropyWithLogits.java
index 4c23683d9ef..1a5011c713a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SoftmaxCrossEntropyWithLogits.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SoftmaxCrossEntropyWithLogits.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes softmax cross entropy cost and gradients to backpropagate.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits.java
index e7cb45231de..c8dcac1a84a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes softmax cross entropy cost and gradients to backpropagate.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Dequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Dequantize.java
index 760dd9fe913..6c9484cab55 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Dequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Dequantize.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.quantization;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -137,7 +136,7 @@ private Options() {
    * @return a new instance of Dequantize
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Dequantize<U> create(Scope scope, Operand<T> input, Operand<TFloat32> minRange, Operand<TFloat32> maxRange, DataType<U> dtype, Options... options) {
+  public static <U extends TNumber, T extends TType> Dequantize<U> create(Scope scope, Operand<T> input, Operand<TFloat32> minRange, Operand<TFloat32> maxRange, Class<U> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Dequantize", scope.makeOpName("Dequantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(minRange.asOutput());
@@ -172,7 +171,7 @@ public static <U extends TNumber, T extends TType> Dequantize<U> create(Scope sc
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Dequantize<TFloat32> create(Scope scope, Operand<T> input, Operand<TFloat32> minRange, Operand<TFloat32> maxRange, Options... options) {
-    return create(scope, input, minRange, maxRange, TFloat32.DTYPE, options);
+    return create(scope, input, minRange, maxRange, TFloat32.class, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Quantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Quantize.java
index 37f58cfa512..61370659157 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Quantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Quantize.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.quantization;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -220,7 +219,7 @@ private Options() {
    * @return a new instance of Quantize
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> Quantize<T> create(Scope scope, Operand<TFloat32> input, Operand<TFloat32> minRange, Operand<TFloat32> maxRange, DataType<T> T, Options... options) {
+  public static <T extends TType> Quantize<T> create(Scope scope, Operand<TFloat32> input, Operand<TFloat32> minRange, Operand<TFloat32> maxRange, Class<T> T, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizeV2", scope.makeOpName("Quantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(minRange.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeAndDequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeAndDequantize.java
index fd75b330e41..c14d9f451d7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeAndDequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeAndDequantize.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Quantizes then dequantizes a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeDownAndShrinkRange.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeDownAndShrinkRange.java
index 362375b40e2..04f13bd833d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeDownAndShrinkRange.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizeDownAndShrinkRange.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.quantization;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -71,7 +70,7 @@ public final class QuantizeDownAndShrinkRange<U extends TType> extends RawOp {
    * @return a new instance of QuantizeDownAndShrinkRange
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> QuantizeDownAndShrinkRange<U> create(Scope scope, Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, DataType<U> outType) {
+  public static <U extends TType, T extends TType> QuantizeDownAndShrinkRange<U> create(Scope scope, Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizeDownAndShrinkRange", scope.makeOpName("QuantizeDownAndShrinkRange"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(inputMin.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndDequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndDequantize.java
index baa20635c51..114d40bd058 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndDequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndDequantize.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.quantization;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -90,7 +89,7 @@ private Options() {
    * @return a new instance of QuantizedMatMulWithBiasAndDequantize
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TNumber, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBiasAndDequantize<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<W> Toutput, Options... options) {
+  public static <W extends TNumber, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBiasAndDequantize<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<W> Toutput, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedMatMulWithBiasAndDequantize", scope.makeOpName("QuantizedMatMulWithBiasAndDequantize"));
     opBuilder.addInput(a.asOutput());
     opBuilder.addInput(b.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndRequantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndRequantize.java
index 950221b1b94..d553cd17e60 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndRequantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/QuantizedMatMulWithBiasAndRequantize.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.quantization;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -89,7 +88,7 @@ private Options() {
    * @return a new instance of QuantizedMatMulWithBiasAndRequantize
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBiasAndRequantize<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, DataType<W> Toutput, Options... options) {
+  public static <W extends TType, T extends TType, U extends TType, V extends TType> QuantizedMatMulWithBiasAndRequantize<W> create(Scope scope, Operand<T> a, Operand<U> b, Operand<V> bias, Operand<TFloat32> minA, Operand<TFloat32> maxA, Operand<TFloat32> minB, Operand<TFloat32> maxB, Operand<TFloat32> minFreezedOutput, Operand<TFloat32> maxFreezedOutput, Class<W> Toutput, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("QuantizedMatMulWithBiasAndRequantize", scope.makeOpName("QuantizedMatMulWithBiasAndRequantize"));
     opBuilder.addInput(a.asOutput());
     opBuilder.addInput(b.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Requantize.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Requantize.java
index 5df8ca0b622..1ee6077abe1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Requantize.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/quantization/Requantize.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.quantization;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -58,7 +57,7 @@ public final class Requantize<U extends TType> extends RawOp {
    * @return a new instance of Requantize
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> Requantize<U> create(Scope scope, Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, Operand<TFloat32> requestedOutputMin, Operand<TFloat32> requestedOutputMax, DataType<U> outType) {
+  public static <U extends TType, T extends TType> Requantize<U> create(Scope scope, Operand<T> input, Operand<TFloat32> inputMin, Operand<TFloat32> inputMax, Operand<TFloat32> requestedOutputMin, Operand<TFloat32> requestedOutputMax, Class<U> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("Requantize", scope.makeOpName("Requantize"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(inputMin.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedBincount.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedBincount.java
index 1e0224aa9ef..b5e54a6a822 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedBincount.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedBincount.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Counts the number of occurrences of each value in an integer array.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCountSparseOutput.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCountSparseOutput.java
index 4829e49488b..5148b8f1f00 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCountSparseOutput.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCountSparseOutput.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs sparse-output bin counting for a ragged tensor input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCross.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCross.java
index 9ea32878257..4a21e8cdfe4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCross.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedCross.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.ragged;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -62,7 +61,7 @@ public final class RaggedCross<T extends TType, U extends TNumber> extends RawOp
    * @return a new instance of RaggedCross
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType, U extends TNumber> RaggedCross<T, U> create(Scope scope, Iterable<Operand<?>> raggedValues, Iterable<Operand<?>> raggedRowSplits, Iterable<Operand<TInt64>> sparseIndices, Iterable<Operand<?>> sparseValues, Iterable<Operand<TInt64>> sparseShape, Iterable<Operand<?>> denseInputs, String inputOrder, Boolean hashedOutput, Long numBuckets, Long hashKey, DataType<T> outValuesType, DataType<U> outRowSplitsType) {
+  public static <T extends TType, U extends TNumber> RaggedCross<T, U> create(Scope scope, Iterable<Operand<?>> raggedValues, Iterable<Operand<?>> raggedRowSplits, Iterable<Operand<TInt64>> sparseIndices, Iterable<Operand<?>> sparseValues, Iterable<Operand<TInt64>> sparseShape, Iterable<Operand<?>> denseInputs, String inputOrder, Boolean hashedOutput, Long numBuckets, Long hashKey, Class<T> outValuesType, Class<U> outRowSplitsType) {
     OperationBuilder opBuilder = scope.env().opBuilder("RaggedCross", scope.makeOpName("RaggedCross"));
     opBuilder.addInputList(Operands.asOutputs(raggedValues));
     opBuilder.addInputList(Operands.asOutputs(raggedRowSplits));
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedRange.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedRange.java
index 9d8f5594fd9..ea23e612ab2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedRange.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedRange.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.ragged;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns a `RaggedTensor` containing the specified sequences of numbers.
@@ -64,7 +62,7 @@ public final class RaggedRange<U extends TNumber, T extends TNumber> extends Raw
    * @return a new instance of RaggedRange
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TNumber> RaggedRange<U, T> create(Scope scope, Operand<T> starts, Operand<T> limits, Operand<T> deltas, DataType<U> Tsplits) {
+  public static <U extends TNumber, T extends TNumber> RaggedRange<U, T> create(Scope scope, Operand<T> starts, Operand<T> limits, Operand<T> deltas, Class<U> Tsplits) {
     OperationBuilder opBuilder = scope.env().opBuilder("RaggedRange", scope.makeOpName("RaggedRange"));
     opBuilder.addInput(starts.asOutput());
     opBuilder.addInput(limits.asOutput());
@@ -85,7 +83,7 @@ public static <U extends TNumber, T extends TNumber> RaggedRange<U, T> create(Sc
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber> RaggedRange<TInt64, T> create(Scope scope, Operand<T> starts, Operand<T> limits, Operand<T> deltas) {
-    return create(scope, starts, limits, deltas, TInt64.DTYPE);
+    return create(scope, starts, limits, deltas, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedTensorFromVariant.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedTensorFromVariant.java
index c9dcfe54bda..cd447134359 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedTensorFromVariant.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/ragged/RaggedTensorFromVariant.java
@@ -19,7 +19,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -67,7 +66,7 @@ public final class RaggedTensorFromVariant<U extends TNumber, T extends TType> e
    * @return a new instance of RaggedTensorFromVariant
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> RaggedTensorFromVariant<U, T> create(Scope scope, Operand<?> encodedRagged, Long inputRaggedRank, Long outputRaggedRank, DataType<T> Tvalues, DataType<U> Tsplits) {
+  public static <U extends TNumber, T extends TType> RaggedTensorFromVariant<U, T> create(Scope scope, Operand<?> encodedRagged, Long inputRaggedRank, Long outputRaggedRank, Class<T> Tvalues, Class<U> Tsplits) {
     OperationBuilder opBuilder = scope.env().opBuilder("RaggedTensorFromVariant", scope.makeOpName("RaggedTensorFromVariant"));
     opBuilder.addInput(encodedRagged.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -91,8 +90,8 @@ public static <U extends TNumber, T extends TType> RaggedTensorFromVariant<U, T>
    * @return a new instance of RaggedTensorFromVariant
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> RaggedTensorFromVariant<TInt64, T> create(Scope scope, Operand<?> encodedRagged, Long inputRaggedRank, Long outputRaggedRank, DataType<T> Tvalues) {
-    return create(scope, encodedRagged, inputRaggedRank, outputRaggedRank, Tvalues, TInt64.DTYPE);
+  public static <T extends TType> RaggedTensorFromVariant<TInt64, T> create(Scope scope, Operand<?> encodedRagged, Long inputRaggedRank, Long outputRaggedRank, Class<T> Tvalues) {
+    return create(scope, encodedRagged, inputRaggedRank, outputRaggedRank, Tvalues, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/Multinomial.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/Multinomial.java
index 58cc57d6c52..fbfb90fba47 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/Multinomial.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/Multinomial.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Draws samples from a multinomial distribution.
@@ -80,7 +78,7 @@ private Options() {
    * @return a new instance of Multinomial
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TNumber> Multinomial<U> create(Scope scope, Operand<T> logits, Operand<TInt32> numSamples, DataType<U> outputDtype, Options... options) {
+  public static <U extends TNumber, T extends TNumber> Multinomial<U> create(Scope scope, Operand<T> logits, Operand<TInt32> numSamples, Class<U> outputDtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("Multinomial", scope.makeOpName("Multinomial"));
     opBuilder.addInput(logits.asOutput());
     opBuilder.addInput(numSamples.asOutput());
@@ -111,7 +109,7 @@ public static <U extends TNumber, T extends TNumber> Multinomial<U> create(Scope
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber> Multinomial<TInt64> create(Scope scope, Operand<T> logits, Operand<TInt32> numSamples, Options... options) {
-    return create(scope, logits, numSamples, TInt64.DTYPE, options);
+    return create(scope, logits, numSamples, TInt64.class, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/NonDeterministicInts.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/NonDeterministicInts.java
index 246974eaf6f..80ee33ef06b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/NonDeterministicInts.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/NonDeterministicInts.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -47,7 +46,7 @@ public final class NonDeterministicInts<U extends TType> extends RawOp implement
    * @return a new instance of NonDeterministicInts
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> NonDeterministicInts<U> create(Scope scope, Operand<T> shape, DataType<U> dtype) {
+  public static <U extends TType, T extends TType> NonDeterministicInts<U> create(Scope scope, Operand<T> shape, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("NonDeterministicInts", scope.makeOpName("NonDeterministicInts"));
     opBuilder.addInput(shape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -64,7 +63,7 @@ public static <U extends TType, T extends TType> NonDeterministicInts<U> create(
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> NonDeterministicInts<TInt64> create(Scope scope, Operand<T> shape) {
-    return create(scope, shape, TInt64.DTYPE);
+    return create(scope, shape, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/ParameterizedTruncatedNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/ParameterizedTruncatedNormal.java
index 4be50b9cde0..c9933d29a17 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/ParameterizedTruncatedNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/ParameterizedTruncatedNormal.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs random values from a normal distribution. The parameters may each be a
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGamma.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGamma.java
index 13963e09ecb..aee60a06932 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGamma.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGamma.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs random values from the Gamma distribution(s) described by alpha.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGammaGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGammaGrad.java
index ce3798cef3f..29c7c625e17 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGammaGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomGammaGrad.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the derivative of a Gamma random sample w.r.t. `alpha`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomPoisson.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomPoisson.java
index d4b516343ac..04da9a55cdd 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomPoisson.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomPoisson.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs random values from the Poisson distribution(s) described by rate.
@@ -91,7 +89,7 @@ private Options() {
    * @return a new instance of RandomPoisson
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TNumber, U extends TNumber> RandomPoisson<V> create(Scope scope, Operand<T> shape, Operand<U> rate, DataType<V> dtype, Options... options) {
+  public static <V extends TNumber, T extends TNumber, U extends TNumber> RandomPoisson<V> create(Scope scope, Operand<T> shape, Operand<U> rate, Class<V> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("RandomPoissonV2", scope.makeOpName("RandomPoisson"));
     opBuilder.addInput(shape.asOutput());
     opBuilder.addInput(rate.asOutput());
@@ -123,7 +121,7 @@ public static <V extends TNumber, T extends TNumber, U extends TNumber> RandomPo
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber, U extends TNumber> RandomPoisson<TInt64> create(Scope scope, Operand<T> shape, Operand<U> rate, Options... options) {
-    return create(scope, shape, rate, TInt64.DTYPE, options);
+    return create(scope, shape, rate, TInt64.class, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomStandardNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomStandardNormal.java
index bdd971cc19d..c9472874e46 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomStandardNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomStandardNormal.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -27,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs random values from a normal distribution.
@@ -79,7 +77,7 @@ private Options() {
    * @return a new instance of RandomStandardNormal
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TNumber> RandomStandardNormal<U> create(Scope scope, Operand<T> shape, DataType<U> dtype, Options... options) {
+  public static <U extends TNumber, T extends TNumber> RandomStandardNormal<U> create(Scope scope, Operand<T> shape, Class<U> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("RandomStandardNormal", scope.makeOpName("RandomStandardNormal"));
     opBuilder.addInput(shape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniform.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniform.java
index 5e42c2d9691..43952690b7d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniform.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniform.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -27,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs random values from a uniform distribution.
@@ -80,7 +78,7 @@ private Options() {
    * @return a new instance of RandomUniform
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TNumber> RandomUniform<U> create(Scope scope, Operand<T> shape, DataType<U> dtype, Options... options) {
+  public static <U extends TNumber, T extends TNumber> RandomUniform<U> create(Scope scope, Operand<T> shape, Class<U> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("RandomUniform", scope.makeOpName("RandomUniform"));
     opBuilder.addInput(shape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniformInt.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniformInt.java
index 5232135ac1c..1b59cf964e7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniformInt.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/RandomUniformInt.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs random integers from a uniform distribution.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulRandomBinomial.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulRandomBinomial.java
index b3c2dfce166..f0d68949af1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulRandomBinomial.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulRandomBinomial.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <V> data type for {@code output()} output
@@ -49,7 +47,7 @@ public final class StatefulRandomBinomial<V extends TNumber> extends RawOp imple
    * @return a new instance of StatefulRandomBinomial
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatefulRandomBinomial<V> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Operand<U> counts, Operand<U> probs, DataType<V> dtype) {
+  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatefulRandomBinomial<V> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Operand<U> counts, Operand<U> probs, Class<V> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatefulRandomBinomial", scope.makeOpName("StatefulRandomBinomial"));
     opBuilder.addInput(resource.asOutput());
     opBuilder.addInput(algorithm.asOutput());
@@ -74,7 +72,7 @@ public static <V extends TNumber, T extends TNumber, U extends TNumber> Stateful
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber, U extends TNumber> StatefulRandomBinomial<TInt64> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Operand<U> counts, Operand<U> probs) {
-    return create(scope, resource, algorithm, shape, counts, probs, TInt64.DTYPE);
+    return create(scope, resource, algorithm, shape, counts, probs, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulStandardNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulStandardNormal.java
index 12aabfa2a73..a01c0f740b8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulStandardNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulStandardNormal.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,7 +50,7 @@ public final class StatefulStandardNormal<U extends TType> extends RawOp impleme
    * @return a new instance of StatefulStandardNormal
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> StatefulStandardNormal<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, DataType<U> dtype) {
+  public static <U extends TType, T extends TType> StatefulStandardNormal<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatefulStandardNormalV2", scope.makeOpName("StatefulStandardNormal"));
     opBuilder.addInput(resource.asOutput());
     opBuilder.addInput(algorithm.asOutput());
@@ -72,7 +71,7 @@ public static <U extends TType, T extends TType> StatefulStandardNormal<U> creat
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> StatefulStandardNormal<TFloat32> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape) {
-    return create(scope, resource, algorithm, shape, TFloat32.DTYPE);
+    return create(scope, resource, algorithm, shape, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulTruncatedNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulTruncatedNormal.java
index 86904de711f..461389eaec5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulTruncatedNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulTruncatedNormal.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -52,7 +51,7 @@ public final class StatefulTruncatedNormal<U extends TType> extends RawOp implem
    * @return a new instance of StatefulTruncatedNormal
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> StatefulTruncatedNormal<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, DataType<U> dtype) {
+  public static <U extends TType, T extends TType> StatefulTruncatedNormal<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatefulTruncatedNormal", scope.makeOpName("StatefulTruncatedNormal"));
     opBuilder.addInput(resource.asOutput());
     opBuilder.addInput(algorithm.asOutput());
@@ -73,7 +72,7 @@ public static <U extends TType, T extends TType> StatefulTruncatedNormal<U> crea
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> StatefulTruncatedNormal<TFloat32> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape) {
-    return create(scope, resource, algorithm, shape, TFloat32.DTYPE);
+    return create(scope, resource, algorithm, shape, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniform.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniform.java
index dd9a2c10af0..89fa8bdd32d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniform.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniform.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -51,7 +50,7 @@ public final class StatefulUniform<U extends TType> extends RawOp implements Ope
    * @return a new instance of StatefulUniform
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> StatefulUniform<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, DataType<U> dtype) {
+  public static <U extends TType, T extends TType> StatefulUniform<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatefulUniform", scope.makeOpName("StatefulUniform"));
     opBuilder.addInput(resource.asOutput());
     opBuilder.addInput(algorithm.asOutput());
@@ -72,7 +71,7 @@ public static <U extends TType, T extends TType> StatefulUniform<U> create(Scope
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> StatefulUniform<TFloat32> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape) {
-    return create(scope, resource, algorithm, shape, TFloat32.DTYPE);
+    return create(scope, resource, algorithm, shape, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniformFullInt.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniformFullInt.java
index 0bf991bd72e..457dcc0b1a8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniformFullInt.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatefulUniformFullInt.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -49,7 +48,7 @@ public final class StatefulUniformFullInt<U extends TType> extends RawOp impleme
    * @return a new instance of StatefulUniformFullInt
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> StatefulUniformFullInt<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, DataType<U> dtype) {
+  public static <U extends TType, T extends TType> StatefulUniformFullInt<U> create(Scope scope, Operand<?> resource, Operand<TInt64> algorithm, Operand<T> shape, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatefulUniformFullInt", scope.makeOpName("StatefulUniformFullInt"));
     opBuilder.addInput(resource.asOutput());
     opBuilder.addInput(algorithm.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessMultinomial.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessMultinomial.java
index 1dbe0e5a7d0..3bc5ab365a9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessMultinomial.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessMultinomial.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Draws samples from a multinomial distribution.
@@ -51,7 +49,7 @@ public final class StatelessMultinomial<V extends TNumber> extends RawOp impleme
    * @return a new instance of StatelessMultinomial
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessMultinomial<V> create(Scope scope, Operand<T> logits, Operand<TInt32> numSamples, Operand<U> seed, DataType<V> outputDtype) {
+  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessMultinomial<V> create(Scope scope, Operand<T> logits, Operand<TInt32> numSamples, Operand<U> seed, Class<V> outputDtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatelessMultinomial", scope.makeOpName("StatelessMultinomial"));
     opBuilder.addInput(logits.asOutput());
     opBuilder.addInput(numSamples.asOutput());
@@ -73,7 +71,7 @@ public static <V extends TNumber, T extends TNumber, U extends TNumber> Stateles
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber, U extends TNumber> StatelessMultinomial<TInt64> create(Scope scope, Operand<T> logits, Operand<TInt32> numSamples, Operand<U> seed) {
-    return create(scope, logits, numSamples, seed, TInt64.DTYPE);
+    return create(scope, logits, numSamples, seed, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessParameterizedTruncatedNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessParameterizedTruncatedNormal.java
index 179160463c7..f422a582a0f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessParameterizedTruncatedNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessParameterizedTruncatedNormal.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * @param <V> data type for {@code output()} output
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomBinomial.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomBinomial.java
index 03543495413..2c2fe32cf12 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomBinomial.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomBinomial.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom random numbers from a binomial distribution.
@@ -55,7 +53,7 @@ public final class StatelessRandomBinomial<W extends TNumber> extends RawOp impl
    * @return a new instance of StatelessRandomBinomial
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TNumber, T extends TNumber, U extends TNumber, V extends TNumber> StatelessRandomBinomial<W> create(Scope scope, Operand<T> shape, Operand<U> seed, Operand<V> counts, Operand<V> probs, DataType<W> dtype) {
+  public static <W extends TNumber, T extends TNumber, U extends TNumber, V extends TNumber> StatelessRandomBinomial<W> create(Scope scope, Operand<T> shape, Operand<U> seed, Operand<V> counts, Operand<V> probs, Class<W> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatelessRandomBinomial", scope.makeOpName("StatelessRandomBinomial"));
     opBuilder.addInput(shape.asOutput());
     opBuilder.addInput(seed.asOutput());
@@ -80,7 +78,7 @@ public static <W extends TNumber, T extends TNumber, U extends TNumber, V extend
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber, U extends TNumber, V extends TNumber> StatelessRandomBinomial<TInt64> create(Scope scope, Operand<T> shape, Operand<U> seed, Operand<V> counts, Operand<V> probs) {
-    return create(scope, shape, seed, counts, probs, TInt64.DTYPE);
+    return create(scope, shape, seed, counts, probs, TInt64.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomGamma.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomGamma.java
index 2c4eef75ff7..7c8aabfa082 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomGamma.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomGamma.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom random numbers from a gamma distribution.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomNormal.java
index 07c5298cca0..2b20afb21b1 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomNormal.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom values from a normal distribution.
@@ -52,7 +50,7 @@ public final class StatelessRandomNormal<V extends TNumber> extends RawOp implem
    * @return a new instance of StatelessRandomNormal
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomNormal<V> create(Scope scope, Operand<T> shape, Operand<U> seed, DataType<V> dtype) {
+  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomNormal<V> create(Scope scope, Operand<T> shape, Operand<U> seed, Class<V> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatelessRandomNormal", scope.makeOpName("StatelessRandomNormal"));
     opBuilder.addInput(shape.asOutput());
     opBuilder.addInput(seed.asOutput());
@@ -71,7 +69,7 @@ public static <V extends TNumber, T extends TNumber, U extends TNumber> Stateles
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber, U extends TNumber> StatelessRandomNormal<TFloat32> create(Scope scope, Operand<T> shape, Operand<U> seed) {
-    return create(scope, shape, seed, TFloat32.DTYPE);
+    return create(scope, shape, seed, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomPoisson.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomPoisson.java
index e71c70e2c1f..cde1f6609b9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomPoisson.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomPoisson.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -27,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom random numbers from a Poisson distribution.
@@ -52,7 +50,7 @@ public final class StatelessRandomPoisson<W extends TNumber> extends RawOp imple
    * @return a new instance of StatelessRandomPoisson
    */
   @Endpoint(describeByClass = true)
-  public static <W extends TNumber, T extends TNumber, U extends TNumber, V extends TNumber> StatelessRandomPoisson<W> create(Scope scope, Operand<T> shape, Operand<U> seed, Operand<V> lam, DataType<W> dtype) {
+  public static <W extends TNumber, T extends TNumber, U extends TNumber, V extends TNumber> StatelessRandomPoisson<W> create(Scope scope, Operand<T> shape, Operand<U> seed, Operand<V> lam, Class<W> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatelessRandomPoisson", scope.makeOpName("StatelessRandomPoisson"));
     opBuilder.addInput(shape.asOutput());
     opBuilder.addInput(seed.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniform.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniform.java
index 9eb6edc67e5..c94282163f6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniform.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniform.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom random values from a uniform distribution.
@@ -53,7 +51,7 @@ public final class StatelessRandomUniform<V extends TNumber> extends RawOp imple
    * @return a new instance of StatelessRandomUniform
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomUniform<V> create(Scope scope, Operand<T> shape, Operand<U> seed, DataType<V> dtype) {
+  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomUniform<V> create(Scope scope, Operand<T> shape, Operand<U> seed, Class<V> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatelessRandomUniform", scope.makeOpName("StatelessRandomUniform"));
     opBuilder.addInput(shape.asOutput());
     opBuilder.addInput(seed.asOutput());
@@ -72,7 +70,7 @@ public static <V extends TNumber, T extends TNumber, U extends TNumber> Stateles
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber, U extends TNumber> StatelessRandomUniform<TFloat32> create(Scope scope, Operand<T> shape, Operand<U> seed) {
-    return create(scope, shape, seed, TFloat32.DTYPE);
+    return create(scope, shape, seed, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformFullInt.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformFullInt.java
index 291cce74d19..b774c165ec0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformFullInt.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformFullInt.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -27,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom random integers from a uniform distribution.
@@ -50,7 +48,7 @@ public final class StatelessRandomUniformFullInt<V extends TNumber> extends RawO
    * @return a new instance of StatelessRandomUniformFullInt
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomUniformFullInt<V> create(Scope scope, Operand<T> shape, Operand<U> seed, DataType<V> dtype) {
+  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessRandomUniformFullInt<V> create(Scope scope, Operand<T> shape, Operand<U> seed, Class<V> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatelessRandomUniformFullInt", scope.makeOpName("StatelessRandomUniformFullInt"));
     opBuilder.addInput(shape.asOutput());
     opBuilder.addInput(seed.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformInt.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformInt.java
index 4695718a186..e8bbba18ab5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformInt.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessRandomUniformInt.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom random integers from a uniform distribution.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessTruncatedNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessTruncatedNormal.java
index 3d4761fc4df..cf260db14a9 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessTruncatedNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/StatelessTruncatedNormal.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -28,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs deterministic pseudorandom values from a truncated normal distribution.
@@ -54,7 +52,7 @@ public final class StatelessTruncatedNormal<V extends TNumber> extends RawOp imp
    * @return a new instance of StatelessTruncatedNormal
    */
   @Endpoint(describeByClass = true)
-  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessTruncatedNormal<V> create(Scope scope, Operand<T> shape, Operand<U> seed, DataType<V> dtype) {
+  public static <V extends TNumber, T extends TNumber, U extends TNumber> StatelessTruncatedNormal<V> create(Scope scope, Operand<T> shape, Operand<U> seed, Class<V> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("StatelessTruncatedNormal", scope.makeOpName("StatelessTruncatedNormal"));
     opBuilder.addInput(shape.asOutput());
     opBuilder.addInput(seed.asOutput());
@@ -73,7 +71,7 @@ public static <V extends TNumber, T extends TNumber, U extends TNumber> Stateles
    */
   @Endpoint(describeByClass = true)
   public static <T extends TNumber, U extends TNumber> StatelessTruncatedNormal<TFloat32> create(Scope scope, Operand<T> shape, Operand<U> seed) {
-    return create(scope, shape, seed, TFloat32.DTYPE);
+    return create(scope, shape, seed, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/TruncatedNormal.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/TruncatedNormal.java
index 7b88d720386..dc7aee67856 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/TruncatedNormal.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/random/TruncatedNormal.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.random;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -27,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs random values from a truncated normal distribution.
@@ -81,7 +79,7 @@ private Options() {
    * @return a new instance of TruncatedNormal
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TNumber> TruncatedNormal<U> create(Scope scope, Operand<T> shape, DataType<U> dtype, Options... options) {
+  public static <U extends TNumber, T extends TNumber> TruncatedNormal<U> create(Scope scope, Operand<T> shape, Class<U> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TruncatedNormal", scope.makeOpName("TruncatedNormal"));
     opBuilder.addInput(shape.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft.java
index 80d3bb85291..b0441b47911 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.signal;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -63,7 +62,7 @@ public final class Irfft<U extends TNumber> extends RawOp implements Operand<U>
    * @return a new instance of Irfft
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Irfft<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, DataType<U> Treal) {
+  public static <U extends TNumber, T extends TType> Irfft<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, Class<U> Treal) {
     OperationBuilder opBuilder = scope.env().opBuilder("IRFFT", scope.makeOpName("Irfft"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(fftLength.asOutput());
@@ -82,7 +81,7 @@ public static <U extends TNumber, T extends TType> Irfft<U> create(Scope scope,
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Irfft<TFloat32> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength) {
-    return create(scope, input, fftLength, TFloat32.DTYPE);
+    return create(scope, input, fftLength, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft2d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft2d.java
index 8acf23a4f23..726e1ea31e6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft2d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft2d.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.signal;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -64,7 +63,7 @@ public final class Irfft2d<U extends TNumber> extends RawOp implements Operand<U
    * @return a new instance of Irfft2d
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Irfft2d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, DataType<U> Treal) {
+  public static <U extends TNumber, T extends TType> Irfft2d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, Class<U> Treal) {
     OperationBuilder opBuilder = scope.env().opBuilder("IRFFT2D", scope.makeOpName("Irfft2d"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(fftLength.asOutput());
@@ -83,7 +82,7 @@ public static <U extends TNumber, T extends TType> Irfft2d<U> create(Scope scope
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Irfft2d<TFloat32> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength) {
-    return create(scope, input, fftLength, TFloat32.DTYPE);
+    return create(scope, input, fftLength, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft3d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft3d.java
index c7b9efabfd2..0d76b5591e7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft3d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Irfft3d.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.signal;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -64,7 +63,7 @@ public final class Irfft3d<U extends TNumber> extends RawOp implements Operand<U
    * @return a new instance of Irfft3d
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TNumber, T extends TType> Irfft3d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, DataType<U> Treal) {
+  public static <U extends TNumber, T extends TType> Irfft3d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, Class<U> Treal) {
     OperationBuilder opBuilder = scope.env().opBuilder("IRFFT3D", scope.makeOpName("Irfft3d"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(fftLength.asOutput());
@@ -83,7 +82,7 @@ public static <U extends TNumber, T extends TType> Irfft3d<U> create(Scope scope
    */
   @Endpoint(describeByClass = true)
   public static <T extends TType> Irfft3d<TFloat32> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength) {
-    return create(scope, input, fftLength, TFloat32.DTYPE);
+    return create(scope, input, fftLength, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft.java
index 9764bcbf0f2..5d03d58a7b7 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.signal;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -59,7 +58,7 @@ public final class Rfft<U extends TType> extends RawOp implements Operand<U> {
    * @return a new instance of Rfft
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TNumber> Rfft<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, DataType<U> Tcomplex) {
+  public static <U extends TType, T extends TNumber> Rfft<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, Class<U> Tcomplex) {
     OperationBuilder opBuilder = scope.env().opBuilder("RFFT", scope.makeOpName("Rfft"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(fftLength.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft2d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft2d.java
index 91187dced7b..ddecd0f133f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft2d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft2d.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.signal;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -60,7 +59,7 @@ public final class Rfft2d<U extends TType> extends RawOp implements Operand<U> {
    * @return a new instance of Rfft2d
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TNumber> Rfft2d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, DataType<U> Tcomplex) {
+  public static <U extends TType, T extends TNumber> Rfft2d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, Class<U> Tcomplex) {
     OperationBuilder opBuilder = scope.env().opBuilder("RFFT2D", scope.makeOpName("Rfft2d"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(fftLength.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft3d.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft3d.java
index 1eb113e9cf1..8fd625822e4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft3d.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/signal/Rfft3d.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.signal;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -60,7 +59,7 @@ public final class Rfft3d<U extends TType> extends RawOp implements Operand<U> {
    * @return a new instance of Rfft3d
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TNumber> Rfft3d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, DataType<U> Tcomplex) {
+  public static <U extends TType, T extends TNumber> Rfft3d<U> create(Scope scope, Operand<T> input, Operand<TInt32> fftLength, Class<U> Tcomplex) {
     OperationBuilder opBuilder = scope.env().opBuilder("RFFT3D", scope.makeOpName("Rfft3d"));
     opBuilder.addInput(input.asOutput());
     opBuilder.addInput(fftLength.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DenseCountSparseOutput.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DenseCountSparseOutput.java
index ed390a7ba47..de6c4ef9eb0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DenseCountSparseOutput.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DenseCountSparseOutput.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs sparse-output bin counting for a tf.tensor input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DeserializeSparse.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DeserializeSparse.java
index a1a6bb2fa49..871910a0874 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DeserializeSparse.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/DeserializeSparse.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -89,7 +88,7 @@ public final class DeserializeSparse<U extends TType> extends RawOp {
    * @return a new instance of DeserializeSparse
    */
   @Endpoint(describeByClass = true)
-  public static <U extends TType, T extends TType> DeserializeSparse<U> create(Scope scope, Operand<T> serializedSparse, DataType<U> dtype) {
+  public static <U extends TType, T extends TType> DeserializeSparse<U> create(Scope scope, Operand<T> serializedSparse, Class<U> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("DeserializeSparse", scope.makeOpName("DeserializeSparse"));
     opBuilder.addInput(serializedSparse.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseAccumulatorTakeGradient.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseAccumulatorTakeGradient.java
index 87cbd112e57..ebd3260f019 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseAccumulatorTakeGradient.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseAccumulatorTakeGradient.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -57,7 +56,7 @@ public final class SparseAccumulatorTakeGradient<T extends TType> extends RawOp
    * @return a new instance of SparseAccumulatorTakeGradient
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> SparseAccumulatorTakeGradient<T> create(Scope scope, Operand<TString> handle, Operand<TInt32> numRequired, DataType<T> dtype) {
+  public static <T extends TType> SparseAccumulatorTakeGradient<T> create(Scope scope, Operand<TString> handle, Operand<TInt32> numRequired, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseAccumulatorTakeGradient", scope.makeOpName("SparseAccumulatorTakeGradient"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(numRequired.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseBincount.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseBincount.java
index 344e27f1346..81d72b8a81e 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseBincount.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseBincount.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Counts the number of occurrences of each value in an integer array.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseConditionalAccumulator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseConditionalAccumulator.java
index b38fc0a6c46..863b2d0b8e3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseConditionalAccumulator.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseConditionalAccumulator.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -92,7 +91,7 @@ private Options() {
    * @return a new instance of SparseConditionalAccumulator
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> SparseConditionalAccumulator create(Scope scope, DataType<T> dtype, Shape shape, Options... options) {
+  public static <T extends TType> SparseConditionalAccumulator create(Scope scope, Class<T> dtype, Shape shape, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseConditionalAccumulator", scope.makeOpName("SparseConditionalAccumulator"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseCountSparseOutput.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseCountSparseOutput.java
index 5e5566db5ec..9f68897c075 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseCountSparseOutput.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseCountSparseOutput.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Performs sparse-output bin counting for a sparse tensor input.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseMatMul.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseMatMul.java
index 295063d560a..fc66887070f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseMatMul.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseMatMul.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Multiply matrix "a" by matrix "b".
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMax.java
index 5288f7b5b88..85c01083363 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMax.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the max of elements across dimensions of a SparseTensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMaxSparse.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMaxSparse.java
index 93ce28bb66a..2c2bbfb9348 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMaxSparse.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseReduceMaxSparse.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the max of elements across dimensions of a SparseTensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMean.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMean.java
index 14dd3df8c2f..72e5784e390 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMean.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMean.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the mean along sparse segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanGrad.java
index b5ae14a0f03..7fc5d505ee6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients for SparseSegmentMean.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanWithNumSegments.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanWithNumSegments.java
index 214d4983c81..9e2047047f8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanWithNumSegments.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentMeanWithNumSegments.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the mean along sparse segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtN.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtN.java
index c1889eae169..dccfcc0d1c2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtN.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtN.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the sum along sparse segments of a tensor divided by the sqrt of N.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNGrad.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNGrad.java
index 3366dbac0c8..a9b7fb3023b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNGrad.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNGrad.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes gradients for SparseSegmentSqrtN.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNWithNumSegments.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNWithNumSegments.java
index 41000f65cdf..127ffa87198 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNWithNumSegments.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSqrtNWithNumSegments.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the sum along sparse segments of a tensor divided by the sqrt of N.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSum.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSum.java
index 991653b2c02..321fbaed9ad 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSum.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSum.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the sum along sparse segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSumWithNumSegments.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSumWithNumSegments.java
index f5414151cbe..830ca794891 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSumWithNumSegments.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSegmentSumWithNumSegments.java
@@ -26,7 +26,6 @@
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Computes the sum along sparse segments of a tensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSoftmax.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSoftmax.java
index 0906213115f..81f0cd9bfe4 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSoftmax.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSoftmax.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Applies softmax to a batched N-D `SparseTensor`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSparseMaximum.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSparseMaximum.java
index 3cbf1a6ef7d..976ac5b4633 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSparseMaximum.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/SparseSparseMaximum.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Returns the element-wise max of two SparseTensors.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/TakeManySparseFromTensorsMap.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/TakeManySparseFromTensorsMap.java
index 7fd5242cb20..6e1bbb6ed27 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/TakeManySparseFromTensorsMap.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/sparse/TakeManySparseFromTensorsMap.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.sparse;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -124,7 +123,7 @@ private Options() {
    * @return a new instance of TakeManySparseFromTensorsMap
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> TakeManySparseFromTensorsMap<T> create(Scope scope, Operand<TInt64> sparseHandles, DataType<T> dtype, Options... options) {
+  public static <T extends TType> TakeManySparseFromTensorsMap<T> create(Scope scope, Operand<TInt64> sparseHandles, Class<T> dtype, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("TakeManySparseFromTensorsMap", scope.makeOpName("TakeManySparseFromTensorsMap"));
     opBuilder.addInput(sparseHandles.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/StringNGrams.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/StringNGrams.java
index 6d68c11fe1b..93d8b60ad87 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/StringNGrams.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/StringNGrams.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Creates ngrams from ragged string data.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/Substr.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/Substr.java
index f1eb5a05485..d09e93458ba 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/Substr.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/Substr.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Return substrings from `Tensor` of strings.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/ToNumber.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/ToNumber.java
index e8e5e4039b9..47ddcfde480 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/ToNumber.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/ToNumber.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.strings;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -29,7 +28,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Converts each string in the input Tensor to the specified numeric type.
@@ -58,7 +56,7 @@ public final class ToNumber<T extends TNumber> extends RawOp implements Operand<
    * @return a new instance of ToNumber
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> ToNumber<T> create(Scope scope, Operand<TString> stringTensor, DataType<T> outType) {
+  public static <T extends TNumber> ToNumber<T> create(Scope scope, Operand<TString> stringTensor, Class<T> outType) {
     OperationBuilder opBuilder = scope.env().opBuilder("StringToNumber", scope.makeOpName("ToNumber"));
     opBuilder.addInput(stringTensor.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -75,7 +73,7 @@ public static <T extends TNumber> ToNumber<T> create(Scope scope, Operand<TStrin
    */
   @Endpoint(describeByClass = true)
   public static ToNumber<TFloat32> create(Scope scope, Operand<TString> stringTensor) {
-    return create(scope, stringTensor, TFloat32.DTYPE);
+    return create(scope, stringTensor, TFloat32.class);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecode.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecode.java
index a39b08b1f5a..1c73739a144 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecode.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecode.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.strings;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -30,7 +29,6 @@
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Decodes each string in `input` into a sequence of Unicode code points.
@@ -116,7 +114,7 @@ private Options() {
    * @return a new instance of UnicodeDecode
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> UnicodeDecode<T> create(Scope scope, Operand<TString> input, String inputEncoding, DataType<T> Tsplits, Options... options) {
+  public static <T extends TNumber> UnicodeDecode<T> create(Scope scope, Operand<TString> input, String inputEncoding, Class<T> Tsplits, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("UnicodeDecode", scope.makeOpName("UnicodeDecode"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -151,7 +149,7 @@ public static <T extends TNumber> UnicodeDecode<T> create(Scope scope, Operand<T
    */
   @Endpoint(describeByClass = true)
   public static UnicodeDecode<TInt64> create(Scope scope, Operand<TString> input, String inputEncoding, Options... options) {
-    return create(scope, input, inputEncoding, TInt64.DTYPE, options);
+    return create(scope, input, inputEncoding, TInt64.class, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecodeWithOffsets.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecodeWithOffsets.java
index ce6977b63ef..f0999df6285 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecodeWithOffsets.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeDecodeWithOffsets.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.strings;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -30,7 +29,6 @@
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Decodes each string in `input` into a sequence of Unicode code points.
@@ -122,7 +120,7 @@ private Options() {
    * @return a new instance of UnicodeDecodeWithOffsets
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TNumber> UnicodeDecodeWithOffsets<T> create(Scope scope, Operand<TString> input, String inputEncoding, DataType<T> Tsplits, Options... options) {
+  public static <T extends TNumber> UnicodeDecodeWithOffsets<T> create(Scope scope, Operand<TString> input, String inputEncoding, Class<T> Tsplits, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("UnicodeDecodeWithOffsets", scope.makeOpName("UnicodeDecodeWithOffsets"));
     opBuilder.addInput(input.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
@@ -157,7 +155,7 @@ public static <T extends TNumber> UnicodeDecodeWithOffsets<T> create(Scope scope
    */
   @Endpoint(describeByClass = true)
   public static UnicodeDecodeWithOffsets<TInt64> create(Scope scope, Operand<TString> input, String inputEncoding, Options... options) {
-    return create(scope, input, inputEncoding, TInt64.DTYPE, options);
+    return create(scope, input, inputEncoding, TInt64.class, options);
   }
   
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeEncode.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeEncode.java
index ab83ef9dcda..e499dee3157 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeEncode.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnicodeEncode.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Encode a tensor of ints into unicode strings.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnsortedSegmentJoin.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnsortedSegmentJoin.java
index df1bbc68b68..2cd28375bd3 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnsortedSegmentJoin.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/strings/UnsortedSegmentJoin.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Joins the elements of `inputs` based on `segment_ids`.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/HistogramSummary.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/HistogramSummary.java
index 669cbf5c2fb..b1bf74ba9d0 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/HistogramSummary.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/HistogramSummary.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs a `Summary` protocol buffer with a histogram.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ImageSummary.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ImageSummary.java
index 5b0f9f8859f..b97fbfcf8da 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ImageSummary.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ImageSummary.java
@@ -21,7 +21,6 @@
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
 import org.tensorflow.Output;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.RawOp;
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
@@ -99,13 +98,13 @@ public Options maxImages(Long maxImages) {
     /**
      * @param badColor Color to use for pixels with non-finite values.
      */
-    public Options badColor(Tensor<?> badColor) {
+    public Options badColor(TType badColor) {
       this.badColor = badColor;
       return this;
     }
     
     private Long maxImages;
-    private Tensor<?> badColor;
+    private TType badColor;
     
     private Options() {
     }
@@ -150,7 +149,7 @@ public static Options maxImages(Long maxImages) {
   /**
    * @param badColor Color to use for pixels with non-finite values.
    */
-  public static Options badColor(Tensor<?> badColor) {
+  public static Options badColor(TType badColor) {
     return new Options().badColor(badColor);
   }
   
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ScalarSummary.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ScalarSummary.java
index 416251aa6b1..20baf84097f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ScalarSummary.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/ScalarSummary.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Outputs a `Summary` protocol buffer with scalar values.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteHistogramSummary.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteHistogramSummary.java
index 2069cefafa4..e460ecc8ca2 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteHistogramSummary.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteHistogramSummary.java
@@ -27,7 +27,6 @@
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  */
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteImageSummary.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteImageSummary.java
index 757ddf59a1c..ff10607477b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteImageSummary.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteImageSummary.java
@@ -28,7 +28,6 @@
 import org.tensorflow.types.TString;
 import org.tensorflow.types.TUint8;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  */
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteScalarSummary.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteScalarSummary.java
index f173651001a..af1291b964f 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteScalarSummary.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/summary/WriteScalarSummary.java
@@ -27,7 +27,6 @@
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  */
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/CrossReplicaSum.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/CrossReplicaSum.java
index 79dc79410a1..d4f982dce61 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/CrossReplicaSum.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/CrossReplicaSum.java
@@ -27,7 +27,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * An Op to sum inputs across replicated TPU instances.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingRaggedTensorBatch.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingRaggedTensorBatch.java
index bf4da86d05d..ea2961fa2b8 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingRaggedTensorBatch.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingRaggedTensorBatch.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Eases the porting of code that uses tf.nn.embedding_lookup().
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseBatch.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseBatch.java
index 2cb7dfb674b..67c4d41ea03 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseBatch.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseBatch.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * An op that enqueues TPUEmbedding input indices from a SparseTensor.
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseTensorBatch.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseTensorBatch.java
index 3d93c6a0f71..df9ea11bf37 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseTensorBatch.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/EnqueueTPUEmbeddingSparseTensorBatch.java
@@ -28,7 +28,6 @@
 import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
 
 /**
  * Eases the porting of code that uses tf.nn.embedding_lookup_sparse().
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeue.java
index ad4d5f52fa8..1713f1f1a4d 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeue.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.tpu;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -45,7 +44,7 @@ public final class InfeedDequeue<T extends TType> extends RawOp implements Opera
    * @return a new instance of InfeedDequeue
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> InfeedDequeue<T> create(Scope scope, DataType<T> dtype, Shape shape) {
+  public static <T extends TType> InfeedDequeue<T> create(Scope scope, Class<T> dtype, Shape shape) {
     OperationBuilder opBuilder = scope.env().opBuilder("InfeedDequeue", scope.makeOpName("InfeedDequeue"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeueTuple.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeueTuple.java
index f471da5b5d7..b90665f21c5 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeueTuple.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/InfeedDequeueTuple.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -46,10 +45,10 @@ public final class InfeedDequeueTuple extends RawOp implements Iterable<Operand<
    * @return a new instance of InfeedDequeueTuple
    */
   @Endpoint(describeByClass = true)
-  public static InfeedDequeueTuple create(Scope scope, List<DataType<?>> dtypes, List<Shape> shapes) {
+  public static InfeedDequeueTuple create(Scope scope, List<Class<? extends TType>> dtypes, List<Shape> shapes) {
     OperationBuilder opBuilder = scope.env().opBuilder("InfeedDequeueTuple", scope.makeOpName("InfeedDequeueTuple"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeue.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeue.java
index 3be811cd039..c11a81f1e54 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeue.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeue.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.tpu;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -69,7 +68,7 @@ private Options() {
    * @return a new instance of OutfeedDequeue
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> OutfeedDequeue<T> create(Scope scope, DataType<T> dtype, Shape shape, Options... options) {
+  public static <T extends TType> OutfeedDequeue<T> create(Scope scope, Class<T> dtype, Shape shape, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OutfeedDequeue", scope.makeOpName("OutfeedDequeue"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeueTuple.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeueTuple.java
index 6b9110232d8..284c2d7a0fb 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeueTuple.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/tpu/OutfeedDequeueTuple.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -71,10 +70,10 @@ private Options() {
    * @return a new instance of OutfeedDequeueTuple
    */
   @Endpoint(describeByClass = true)
-  public static OutfeedDequeueTuple create(Scope scope, List<DataType<?>> dtypes, List<Shape> shapes, Options... options) {
+  public static OutfeedDequeueTuple create(Scope scope, List<Class<? extends TType>> dtypes, List<Shape> shapes, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("OutfeedDequeueTuple", scope.makeOpName("OutfeedDequeueTuple"));
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/AccumulatorTakeGradient.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/AccumulatorTakeGradient.java
index 51de5485845..c5371e853c6 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/AccumulatorTakeGradient.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/AccumulatorTakeGradient.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.train;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -55,7 +54,7 @@ public final class AccumulatorTakeGradient<T extends TType> extends RawOp implem
    * @return a new instance of AccumulatorTakeGradient
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> AccumulatorTakeGradient<T> create(Scope scope, Operand<TString> handle, Operand<TInt32> numRequired, DataType<T> dtype) {
+  public static <T extends TType> AccumulatorTakeGradient<T> create(Scope scope, Operand<TString> handle, Operand<TInt32> numRequired, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("AccumulatorTakeGradient", scope.makeOpName("AccumulatorTakeGradient"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(numRequired.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ConditionalAccumulator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ConditionalAccumulator.java
index f35e9f5001f..aaf8a6f767a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ConditionalAccumulator.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ConditionalAccumulator.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.train;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -92,7 +91,7 @@ private Options() {
    * @return a new instance of ConditionalAccumulator
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> ConditionalAccumulator create(Scope scope, DataType<T> dtype, Shape shape, Options... options) {
+  public static <T extends TType> ConditionalAccumulator create(Scope scope, Class<T> dtype, Shape shape, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ConditionalAccumulator", scope.makeOpName("ConditionalAccumulator"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceAccumulatorTakeGradient.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceAccumulatorTakeGradient.java
index c27fb1db524..75601de5579 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceAccumulatorTakeGradient.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceAccumulatorTakeGradient.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.train;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -53,7 +52,7 @@ public final class ResourceAccumulatorTakeGradient<T extends TType> extends RawO
    * @return a new instance of ResourceAccumulatorTakeGradient
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> ResourceAccumulatorTakeGradient<T> create(Scope scope, Operand<?> handle, Operand<TInt32> numRequired, DataType<T> dtype) {
+  public static <T extends TType> ResourceAccumulatorTakeGradient<T> create(Scope scope, Operand<?> handle, Operand<TInt32> numRequired, Class<T> dtype) {
     OperationBuilder opBuilder = scope.env().opBuilder("ResourceAccumulatorTakeGradient", scope.makeOpName("ResourceAccumulatorTakeGradient"));
     opBuilder.addInput(handle.asOutput());
     opBuilder.addInput(numRequired.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceConditionalAccumulator.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceConditionalAccumulator.java
index 2935d3ca6bc..9bc3453a72a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceConditionalAccumulator.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/ResourceConditionalAccumulator.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.train;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -92,7 +91,7 @@ private Options() {
    * @return a new instance of ResourceConditionalAccumulator
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> ResourceConditionalAccumulator create(Scope scope, DataType<T> dtype, Shape shape, Options... options) {
+  public static <T extends TType> ResourceConditionalAccumulator create(Scope scope, Class<T> dtype, Shape shape, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("ResourceConditionalAccumulator", scope.makeOpName("ResourceConditionalAccumulator"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/Restore.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/Restore.java
index 5a2466f501d..721e40e058a 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/Restore.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/Restore.java
@@ -20,7 +20,6 @@
 import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -65,13 +64,13 @@ public final class Restore extends RawOp implements Iterable<Operand<TType>> {
    * @return a new instance of Restore
    */
   @Endpoint(describeByClass = true)
-  public static Restore create(Scope scope, Operand<TString> prefix, Operand<TString> tensorNames, Operand<TString> shapeAndSlices, List<DataType<?>> dtypes) {
+  public static Restore create(Scope scope, Operand<TString> prefix, Operand<TString> tensorNames, Operand<TString> shapeAndSlices, List<Class<? extends TType>> dtypes) {
     OperationBuilder opBuilder = scope.env().opBuilder("RestoreV2", scope.makeOpName("Restore"));
     opBuilder.addInput(prefix.asOutput());
     opBuilder.addInput(tensorNames.asOutput());
     opBuilder.addInput(shapeAndSlices.asOutput());
     opBuilder = scope.applyControlDependencies(opBuilder);
-    DataType[] dtypesArray = new DataType[dtypes.size()];
+    Class[] dtypesArray = new Class[dtypes.size()];
     for (int i = 0; i < dtypesArray.length; ++i) {
       dtypesArray[i] = dtypes.get(i);
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/RestoreSlice.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/RestoreSlice.java
index 631a127a44f..24b13ac30af 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/RestoreSlice.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/train/RestoreSlice.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.train;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -79,7 +78,7 @@ private Options() {
    * @return a new instance of RestoreSlice
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> RestoreSlice<T> create(Scope scope, Operand<TString> filePattern, Operand<TString> tensorName, Operand<TString> shapeAndSlice, DataType<T> dt, Options... options) {
+  public static <T extends TType> RestoreSlice<T> create(Scope scope, Operand<TString> filePattern, Operand<TString> tensorName, Operand<TString> shapeAndSlice, Class<T> dt, Options... options) {
     OperationBuilder opBuilder = scope.env().opBuilder("RestoreSlice", scope.makeOpName("RestoreSlice"));
     opBuilder.addInput(filePattern.asOutput());
     opBuilder.addInput(tensorName.asOutput());
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/xla/Recv.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/xla/Recv.java
index fe26ee27587..63cdcc2767b 100644
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/xla/Recv.java
+++ b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/xla/Recv.java
@@ -17,7 +17,6 @@
 
 package org.tensorflow.op.xla;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.OperationBuilder;
@@ -50,7 +49,7 @@ public final class Recv<T extends TType> extends RawOp implements Operand<T> {
    * @return a new instance of Recv
    */
   @Endpoint(describeByClass = true)
-  public static <T extends TType> Recv<T> create(Scope scope, DataType<T> dtype, String tensorName, Shape shape) {
+  public static <T extends TType> Recv<T> create(Scope scope, Class<T> dtype, String tensorName, Shape shape) {
     OperationBuilder opBuilder = scope.env().opBuilder("XlaRecv", scope.makeOpName("Recv"));
     opBuilder = scope.applyControlDependencies(opBuilder);
     opBuilder.setAttr("dtype", dtype);
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Device.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Device.java
deleted file mode 100644
index 59507a00680..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Device.java
+++ /dev/null
@@ -1,977 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-/**
- * <pre>
- * A 'device' is a physical entity in the system and is comprised of several
- * resources.
- * </pre>
- *
- * Protobuf type {@code tensorflow.profiler.Device}
- */
-public  final class Device extends
-    com.google.protobuf.GeneratedMessageV3 implements
-    // @@protoc_insertion_point(message_implements:tensorflow.profiler.Device)
-    DeviceOrBuilder {
-private static final long serialVersionUID = 0L;
-  // Use Device.newBuilder() to construct.
-  private Device(com.google.protobuf.GeneratedMessageV3.Builder<?> builder) {
-    super(builder);
-  }
-  private Device() {
-    name_ = "";
-  }
-
-  @java.lang.Override
-  @SuppressWarnings({"unused"})
-  protected java.lang.Object newInstance(
-      UnusedPrivateParameter unused) {
-    return new Device();
-  }
-
-  @java.lang.Override
-  public final com.google.protobuf.UnknownFieldSet
-  getUnknownFields() {
-    return this.unknownFields;
-  }
-  private Device(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    this();
-    if (extensionRegistry == null) {
-      throw new java.lang.NullPointerException();
-    }
-    int mutable_bitField0_ = 0;
-    com.google.protobuf.UnknownFieldSet.Builder unknownFields =
-        com.google.protobuf.UnknownFieldSet.newBuilder();
-    try {
-      boolean done = false;
-      while (!done) {
-        int tag = input.readTag();
-        switch (tag) {
-          case 0:
-            done = true;
-            break;
-          case 10: {
-            java.lang.String s = input.readStringRequireUtf8();
-
-            name_ = s;
-            break;
-          }
-          case 16: {
-
-            deviceId_ = input.readUInt32();
-            break;
-          }
-          case 26: {
-            if (!((mutable_bitField0_ & 0x00000001) != 0)) {
-              resources_ = com.google.protobuf.MapField.newMapField(
-                  ResourcesDefaultEntryHolder.defaultEntry);
-              mutable_bitField0_ |= 0x00000001;
-            }
-            com.google.protobuf.MapEntry<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-            resources__ = input.readMessage(
-                ResourcesDefaultEntryHolder.defaultEntry.getParserForType(), extensionRegistry);
-            resources_.getMutableMap().put(
-                resources__.getKey(), resources__.getValue());
-            break;
-          }
-          default: {
-            if (!parseUnknownField(
-                input, unknownFields, extensionRegistry, tag)) {
-              done = true;
-            }
-            break;
-          }
-        }
-      }
-    } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-      throw e.setUnfinishedMessage(this);
-    } catch (java.io.IOException e) {
-      throw new com.google.protobuf.InvalidProtocolBufferException(
-          e).setUnfinishedMessage(this);
-    } finally {
-      this.unknownFields = unknownFields.build();
-      makeExtensionsImmutable();
-    }
-  }
-  public static final com.google.protobuf.Descriptors.Descriptor
-      getDescriptor() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Device_descriptor;
-  }
-
-  @SuppressWarnings({"rawtypes"})
-  @java.lang.Override
-  protected com.google.protobuf.MapField internalGetMapField(
-      int number) {
-    switch (number) {
-      case 3:
-        return internalGetResources();
-      default:
-        throw new RuntimeException(
-            "Invalid map field number: " + number);
-    }
-  }
-  @java.lang.Override
-  protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internalGetFieldAccessorTable() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Device_fieldAccessorTable
-        .ensureFieldAccessorsInitialized(
-            org.tensorflow.proto.framework.Device.class, org.tensorflow.proto.framework.Device.Builder.class);
-  }
-
-  public static final int NAME_FIELD_NUMBER = 1;
-  private volatile java.lang.Object name_;
-  /**
-   * <pre>
-   * The name of the device.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  public java.lang.String getName() {
-    java.lang.Object ref = name_;
-    if (ref instanceof java.lang.String) {
-      return (java.lang.String) ref;
-    } else {
-      com.google.protobuf.ByteString bs = 
-          (com.google.protobuf.ByteString) ref;
-      java.lang.String s = bs.toStringUtf8();
-      name_ = s;
-      return s;
-    }
-  }
-  /**
-   * <pre>
-   * The name of the device.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  public com.google.protobuf.ByteString
-      getNameBytes() {
-    java.lang.Object ref = name_;
-    if (ref instanceof java.lang.String) {
-      com.google.protobuf.ByteString b = 
-          com.google.protobuf.ByteString.copyFromUtf8(
-              (java.lang.String) ref);
-      name_ = b;
-      return b;
-    } else {
-      return (com.google.protobuf.ByteString) ref;
-    }
-  }
-
-  public static final int DEVICE_ID_FIELD_NUMBER = 2;
-  private int deviceId_;
-  /**
-   * <pre>
-   * The id of this device, unique in a single trace.
-   * </pre>
-   *
-   * <code>uint32 device_id = 2;</code>
-   */
-  public int getDeviceId() {
-    return deviceId_;
-  }
-
-  public static final int RESOURCES_FIELD_NUMBER = 3;
-  private static final class ResourcesDefaultEntryHolder {
-    static final com.google.protobuf.MapEntry<
-        java.lang.Integer, org.tensorflow.proto.framework.Resource> defaultEntry =
-            com.google.protobuf.MapEntry
-            .<java.lang.Integer, org.tensorflow.proto.framework.Resource>newDefaultInstance(
-                org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Device_ResourcesEntry_descriptor, 
-                com.google.protobuf.WireFormat.FieldType.UINT32,
-                0,
-                com.google.protobuf.WireFormat.FieldType.MESSAGE,
-                org.tensorflow.proto.framework.Resource.getDefaultInstance());
-  }
-  private com.google.protobuf.MapField<
-      java.lang.Integer, org.tensorflow.proto.framework.Resource> resources_;
-  private com.google.protobuf.MapField<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-  internalGetResources() {
-    if (resources_ == null) {
-      return com.google.protobuf.MapField.emptyMapField(
-          ResourcesDefaultEntryHolder.defaultEntry);
-    }
-    return resources_;
-  }
-
-  public int getResourcesCount() {
-    return internalGetResources().getMap().size();
-  }
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-
-  public boolean containsResources(
-      int key) {
-    
-    return internalGetResources().getMap().containsKey(key);
-  }
-  /**
-   * Use {@link #getResourcesMap()} instead.
-   */
-  @java.lang.Deprecated
-  public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> getResources() {
-    return getResourcesMap();
-  }
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-
-  public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> getResourcesMap() {
-    return internalGetResources().getMap();
-  }
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-
-  public org.tensorflow.proto.framework.Resource getResourcesOrDefault(
-      int key,
-      org.tensorflow.proto.framework.Resource defaultValue) {
-    
-    java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> map =
-        internalGetResources().getMap();
-    return map.containsKey(key) ? map.get(key) : defaultValue;
-  }
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-
-  public org.tensorflow.proto.framework.Resource getResourcesOrThrow(
-      int key) {
-    
-    java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> map =
-        internalGetResources().getMap();
-    if (!map.containsKey(key)) {
-      throw new java.lang.IllegalArgumentException();
-    }
-    return map.get(key);
-  }
-
-  private byte memoizedIsInitialized = -1;
-  @java.lang.Override
-  public final boolean isInitialized() {
-    byte isInitialized = memoizedIsInitialized;
-    if (isInitialized == 1) return true;
-    if (isInitialized == 0) return false;
-
-    memoizedIsInitialized = 1;
-    return true;
-  }
-
-  @java.lang.Override
-  public void writeTo(com.google.protobuf.CodedOutputStream output)
-                      throws java.io.IOException {
-    if (!getNameBytes().isEmpty()) {
-      com.google.protobuf.GeneratedMessageV3.writeString(output, 1, name_);
-    }
-    if (deviceId_ != 0) {
-      output.writeUInt32(2, deviceId_);
-    }
-    com.google.protobuf.GeneratedMessageV3
-      .serializeIntegerMapTo(
-        output,
-        internalGetResources(),
-        ResourcesDefaultEntryHolder.defaultEntry,
-        3);
-    unknownFields.writeTo(output);
-  }
-
-  @java.lang.Override
-  public int getSerializedSize() {
-    int size = memoizedSize;
-    if (size != -1) return size;
-
-    size = 0;
-    if (!getNameBytes().isEmpty()) {
-      size += com.google.protobuf.GeneratedMessageV3.computeStringSize(1, name_);
-    }
-    if (deviceId_ != 0) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeUInt32Size(2, deviceId_);
-    }
-    for (java.util.Map.Entry<java.lang.Integer, org.tensorflow.proto.framework.Resource> entry
-         : internalGetResources().getMap().entrySet()) {
-      com.google.protobuf.MapEntry<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-      resources__ = ResourcesDefaultEntryHolder.defaultEntry.newBuilderForType()
-          .setKey(entry.getKey())
-          .setValue(entry.getValue())
-          .build();
-      size += com.google.protobuf.CodedOutputStream
-          .computeMessageSize(3, resources__);
-    }
-    size += unknownFields.getSerializedSize();
-    memoizedSize = size;
-    return size;
-  }
-
-  @java.lang.Override
-  public boolean equals(final java.lang.Object obj) {
-    if (obj == this) {
-     return true;
-    }
-    if (!(obj instanceof org.tensorflow.proto.framework.Device)) {
-      return super.equals(obj);
-    }
-    org.tensorflow.proto.framework.Device other = (org.tensorflow.proto.framework.Device) obj;
-
-    if (!getName()
-        .equals(other.getName())) return false;
-    if (getDeviceId()
-        != other.getDeviceId()) return false;
-    if (!internalGetResources().equals(
-        other.internalGetResources())) return false;
-    if (!unknownFields.equals(other.unknownFields)) return false;
-    return true;
-  }
-
-  @java.lang.Override
-  public int hashCode() {
-    if (memoizedHashCode != 0) {
-      return memoizedHashCode;
-    }
-    int hash = 41;
-    hash = (19 * hash) + getDescriptor().hashCode();
-    hash = (37 * hash) + NAME_FIELD_NUMBER;
-    hash = (53 * hash) + getName().hashCode();
-    hash = (37 * hash) + DEVICE_ID_FIELD_NUMBER;
-    hash = (53 * hash) + getDeviceId();
-    if (!internalGetResources().getMap().isEmpty()) {
-      hash = (37 * hash) + RESOURCES_FIELD_NUMBER;
-      hash = (53 * hash) + internalGetResources().hashCode();
-    }
-    hash = (29 * hash) + unknownFields.hashCode();
-    memoizedHashCode = hash;
-    return hash;
-  }
-
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      java.nio.ByteBuffer data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      java.nio.ByteBuffer data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      com.google.protobuf.ByteString data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      com.google.protobuf.ByteString data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(byte[] data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      byte[] data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Device parseDelimitedFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Device parseDelimitedFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      com.google.protobuf.CodedInputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Device parseFrom(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-
-  @java.lang.Override
-  public Builder newBuilderForType() { return newBuilder(); }
-  public static Builder newBuilder() {
-    return DEFAULT_INSTANCE.toBuilder();
-  }
-  public static Builder newBuilder(org.tensorflow.proto.framework.Device prototype) {
-    return DEFAULT_INSTANCE.toBuilder().mergeFrom(prototype);
-  }
-  @java.lang.Override
-  public Builder toBuilder() {
-    return this == DEFAULT_INSTANCE
-        ? new Builder() : new Builder().mergeFrom(this);
-  }
-
-  @java.lang.Override
-  protected Builder newBuilderForType(
-      com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-    Builder builder = new Builder(parent);
-    return builder;
-  }
-  /**
-   * <pre>
-   * A 'device' is a physical entity in the system and is comprised of several
-   * resources.
-   * </pre>
-   *
-   * Protobuf type {@code tensorflow.profiler.Device}
-   */
-  public static final class Builder extends
-      com.google.protobuf.GeneratedMessageV3.Builder<Builder> implements
-      // @@protoc_insertion_point(builder_implements:tensorflow.profiler.Device)
-      org.tensorflow.proto.framework.DeviceOrBuilder {
-    public static final com.google.protobuf.Descriptors.Descriptor
-        getDescriptor() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Device_descriptor;
-    }
-
-    @SuppressWarnings({"rawtypes"})
-    protected com.google.protobuf.MapField internalGetMapField(
-        int number) {
-      switch (number) {
-        case 3:
-          return internalGetResources();
-        default:
-          throw new RuntimeException(
-              "Invalid map field number: " + number);
-      }
-    }
-    @SuppressWarnings({"rawtypes"})
-    protected com.google.protobuf.MapField internalGetMutableMapField(
-        int number) {
-      switch (number) {
-        case 3:
-          return internalGetMutableResources();
-        default:
-          throw new RuntimeException(
-              "Invalid map field number: " + number);
-      }
-    }
-    @java.lang.Override
-    protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-        internalGetFieldAccessorTable() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Device_fieldAccessorTable
-          .ensureFieldAccessorsInitialized(
-              org.tensorflow.proto.framework.Device.class, org.tensorflow.proto.framework.Device.Builder.class);
-    }
-
-    // Construct using org.tensorflow.proto.framework.Device.newBuilder()
-    private Builder() {
-      maybeForceBuilderInitialization();
-    }
-
-    private Builder(
-        com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-      super(parent);
-      maybeForceBuilderInitialization();
-    }
-    private void maybeForceBuilderInitialization() {
-      if (com.google.protobuf.GeneratedMessageV3
-              .alwaysUseFieldBuilders) {
-      }
-    }
-    @java.lang.Override
-    public Builder clear() {
-      super.clear();
-      name_ = "";
-
-      deviceId_ = 0;
-
-      internalGetMutableResources().clear();
-      return this;
-    }
-
-    @java.lang.Override
-    public com.google.protobuf.Descriptors.Descriptor
-        getDescriptorForType() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Device_descriptor;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Device getDefaultInstanceForType() {
-      return org.tensorflow.proto.framework.Device.getDefaultInstance();
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Device build() {
-      org.tensorflow.proto.framework.Device result = buildPartial();
-      if (!result.isInitialized()) {
-        throw newUninitializedMessageException(result);
-      }
-      return result;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Device buildPartial() {
-      org.tensorflow.proto.framework.Device result = new org.tensorflow.proto.framework.Device(this);
-      int from_bitField0_ = bitField0_;
-      result.name_ = name_;
-      result.deviceId_ = deviceId_;
-      result.resources_ = internalGetResources();
-      result.resources_.makeImmutable();
-      onBuilt();
-      return result;
-    }
-
-    @java.lang.Override
-    public Builder clone() {
-      return super.clone();
-    }
-    @java.lang.Override
-    public Builder setField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.setField(field, value);
-    }
-    @java.lang.Override
-    public Builder clearField(
-        com.google.protobuf.Descriptors.FieldDescriptor field) {
-      return super.clearField(field);
-    }
-    @java.lang.Override
-    public Builder clearOneof(
-        com.google.protobuf.Descriptors.OneofDescriptor oneof) {
-      return super.clearOneof(oneof);
-    }
-    @java.lang.Override
-    public Builder setRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        int index, java.lang.Object value) {
-      return super.setRepeatedField(field, index, value);
-    }
-    @java.lang.Override
-    public Builder addRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.addRepeatedField(field, value);
-    }
-    @java.lang.Override
-    public Builder mergeFrom(com.google.protobuf.Message other) {
-      if (other instanceof org.tensorflow.proto.framework.Device) {
-        return mergeFrom((org.tensorflow.proto.framework.Device)other);
-      } else {
-        super.mergeFrom(other);
-        return this;
-      }
-    }
-
-    public Builder mergeFrom(org.tensorflow.proto.framework.Device other) {
-      if (other == org.tensorflow.proto.framework.Device.getDefaultInstance()) return this;
-      if (!other.getName().isEmpty()) {
-        name_ = other.name_;
-        onChanged();
-      }
-      if (other.getDeviceId() != 0) {
-        setDeviceId(other.getDeviceId());
-      }
-      internalGetMutableResources().mergeFrom(
-          other.internalGetResources());
-      this.mergeUnknownFields(other.unknownFields);
-      onChanged();
-      return this;
-    }
-
-    @java.lang.Override
-    public final boolean isInitialized() {
-      return true;
-    }
-
-    @java.lang.Override
-    public Builder mergeFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws java.io.IOException {
-      org.tensorflow.proto.framework.Device parsedMessage = null;
-      try {
-        parsedMessage = PARSER.parsePartialFrom(input, extensionRegistry);
-      } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-        parsedMessage = (org.tensorflow.proto.framework.Device) e.getUnfinishedMessage();
-        throw e.unwrapIOException();
-      } finally {
-        if (parsedMessage != null) {
-          mergeFrom(parsedMessage);
-        }
-      }
-      return this;
-    }
-    private int bitField0_;
-
-    private java.lang.Object name_ = "";
-    /**
-     * <pre>
-     * The name of the device.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public java.lang.String getName() {
-      java.lang.Object ref = name_;
-      if (!(ref instanceof java.lang.String)) {
-        com.google.protobuf.ByteString bs =
-            (com.google.protobuf.ByteString) ref;
-        java.lang.String s = bs.toStringUtf8();
-        name_ = s;
-        return s;
-      } else {
-        return (java.lang.String) ref;
-      }
-    }
-    /**
-     * <pre>
-     * The name of the device.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public com.google.protobuf.ByteString
-        getNameBytes() {
-      java.lang.Object ref = name_;
-      if (ref instanceof String) {
-        com.google.protobuf.ByteString b = 
-            com.google.protobuf.ByteString.copyFromUtf8(
-                (java.lang.String) ref);
-        name_ = b;
-        return b;
-      } else {
-        return (com.google.protobuf.ByteString) ref;
-      }
-    }
-    /**
-     * <pre>
-     * The name of the device.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public Builder setName(
-        java.lang.String value) {
-      if (value == null) {
-    throw new NullPointerException();
-  }
-  
-      name_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The name of the device.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public Builder clearName() {
-      
-      name_ = getDefaultInstance().getName();
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The name of the device.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public Builder setNameBytes(
-        com.google.protobuf.ByteString value) {
-      if (value == null) {
-    throw new NullPointerException();
-  }
-  checkByteStringIsUtf8(value);
-      
-      name_ = value;
-      onChanged();
-      return this;
-    }
-
-    private int deviceId_ ;
-    /**
-     * <pre>
-     * The id of this device, unique in a single trace.
-     * </pre>
-     *
-     * <code>uint32 device_id = 2;</code>
-     */
-    public int getDeviceId() {
-      return deviceId_;
-    }
-    /**
-     * <pre>
-     * The id of this device, unique in a single trace.
-     * </pre>
-     *
-     * <code>uint32 device_id = 2;</code>
-     */
-    public Builder setDeviceId(int value) {
-      
-      deviceId_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The id of this device, unique in a single trace.
-     * </pre>
-     *
-     * <code>uint32 device_id = 2;</code>
-     */
-    public Builder clearDeviceId() {
-      
-      deviceId_ = 0;
-      onChanged();
-      return this;
-    }
-
-    private com.google.protobuf.MapField<
-        java.lang.Integer, org.tensorflow.proto.framework.Resource> resources_;
-    private com.google.protobuf.MapField<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-    internalGetResources() {
-      if (resources_ == null) {
-        return com.google.protobuf.MapField.emptyMapField(
-            ResourcesDefaultEntryHolder.defaultEntry);
-      }
-      return resources_;
-    }
-    private com.google.protobuf.MapField<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-    internalGetMutableResources() {
-      onChanged();;
-      if (resources_ == null) {
-        resources_ = com.google.protobuf.MapField.newMapField(
-            ResourcesDefaultEntryHolder.defaultEntry);
-      }
-      if (!resources_.isMutable()) {
-        resources_ = resources_.copy();
-      }
-      return resources_;
-    }
-
-    public int getResourcesCount() {
-      return internalGetResources().getMap().size();
-    }
-    /**
-     * <pre>
-     * The resources on this device, keyed by resource_id;
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-     */
-
-    public boolean containsResources(
-        int key) {
-      
-      return internalGetResources().getMap().containsKey(key);
-    }
-    /**
-     * Use {@link #getResourcesMap()} instead.
-     */
-    @java.lang.Deprecated
-    public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> getResources() {
-      return getResourcesMap();
-    }
-    /**
-     * <pre>
-     * The resources on this device, keyed by resource_id;
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-     */
-
-    public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> getResourcesMap() {
-      return internalGetResources().getMap();
-    }
-    /**
-     * <pre>
-     * The resources on this device, keyed by resource_id;
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-     */
-
-    public org.tensorflow.proto.framework.Resource getResourcesOrDefault(
-        int key,
-        org.tensorflow.proto.framework.Resource defaultValue) {
-      
-      java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> map =
-          internalGetResources().getMap();
-      return map.containsKey(key) ? map.get(key) : defaultValue;
-    }
-    /**
-     * <pre>
-     * The resources on this device, keyed by resource_id;
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-     */
-
-    public org.tensorflow.proto.framework.Resource getResourcesOrThrow(
-        int key) {
-      
-      java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> map =
-          internalGetResources().getMap();
-      if (!map.containsKey(key)) {
-        throw new java.lang.IllegalArgumentException();
-      }
-      return map.get(key);
-    }
-
-    public Builder clearResources() {
-      internalGetMutableResources().getMutableMap()
-          .clear();
-      return this;
-    }
-    /**
-     * <pre>
-     * The resources on this device, keyed by resource_id;
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-     */
-
-    public Builder removeResources(
-        int key) {
-      
-      internalGetMutableResources().getMutableMap()
-          .remove(key);
-      return this;
-    }
-    /**
-     * Use alternate mutation accessors instead.
-     */
-    @java.lang.Deprecated
-    public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-    getMutableResources() {
-      return internalGetMutableResources().getMutableMap();
-    }
-    /**
-     * <pre>
-     * The resources on this device, keyed by resource_id;
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-     */
-    public Builder putResources(
-        int key,
-        org.tensorflow.proto.framework.Resource value) {
-      
-      if (value == null) { throw new java.lang.NullPointerException(); }
-      internalGetMutableResources().getMutableMap()
-          .put(key, value);
-      return this;
-    }
-    /**
-     * <pre>
-     * The resources on this device, keyed by resource_id;
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-     */
-
-    public Builder putAllResources(
-        java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource> values) {
-      internalGetMutableResources().getMutableMap()
-          .putAll(values);
-      return this;
-    }
-    @java.lang.Override
-    public final Builder setUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.setUnknownFields(unknownFields);
-    }
-
-    @java.lang.Override
-    public final Builder mergeUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.mergeUnknownFields(unknownFields);
-    }
-
-
-    // @@protoc_insertion_point(builder_scope:tensorflow.profiler.Device)
-  }
-
-  // @@protoc_insertion_point(class_scope:tensorflow.profiler.Device)
-  private static final org.tensorflow.proto.framework.Device DEFAULT_INSTANCE;
-  static {
-    DEFAULT_INSTANCE = new org.tensorflow.proto.framework.Device();
-  }
-
-  public static org.tensorflow.proto.framework.Device getDefaultInstance() {
-    return DEFAULT_INSTANCE;
-  }
-
-  private static final com.google.protobuf.Parser<Device>
-      PARSER = new com.google.protobuf.AbstractParser<Device>() {
-    @java.lang.Override
-    public Device parsePartialFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws com.google.protobuf.InvalidProtocolBufferException {
-      return new Device(input, extensionRegistry);
-    }
-  };
-
-  public static com.google.protobuf.Parser<Device> parser() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public com.google.protobuf.Parser<Device> getParserForType() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public org.tensorflow.proto.framework.Device getDefaultInstanceForType() {
-    return DEFAULT_INSTANCE;
-  }
-
-}
-
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/DeviceOrBuilder.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/DeviceOrBuilder.java
deleted file mode 100644
index 3ce193459ff..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/DeviceOrBuilder.java
+++ /dev/null
@@ -1,90 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-public interface DeviceOrBuilder extends
-    // @@protoc_insertion_point(interface_extends:tensorflow.profiler.Device)
-    com.google.protobuf.MessageOrBuilder {
-
-  /**
-   * <pre>
-   * The name of the device.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  java.lang.String getName();
-  /**
-   * <pre>
-   * The name of the device.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  com.google.protobuf.ByteString
-      getNameBytes();
-
-  /**
-   * <pre>
-   * The id of this device, unique in a single trace.
-   * </pre>
-   *
-   * <code>uint32 device_id = 2;</code>
-   */
-  int getDeviceId();
-
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-  int getResourcesCount();
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-  boolean containsResources(
-      int key);
-  /**
-   * Use {@link #getResourcesMap()} instead.
-   */
-  @java.lang.Deprecated
-  java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-  getResources();
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-  java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Resource>
-  getResourcesMap();
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-
-  org.tensorflow.proto.framework.Resource getResourcesOrDefault(
-      int key,
-      org.tensorflow.proto.framework.Resource defaultValue);
-  /**
-   * <pre>
-   * The resources on this device, keyed by resource_id;
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Resource&gt; resources = 3;</code>
-   */
-
-  org.tensorflow.proto.framework.Resource getResourcesOrThrow(
-      int key);
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Resource.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Resource.java
deleted file mode 100644
index 7743aecf24e..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Resource.java
+++ /dev/null
@@ -1,659 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-/**
- * <pre>
- * A 'resource' generally is a specific computation component on a device. These
- * can range from threads on CPUs to specific arithmetic units on hardware
- * devices.
- * </pre>
- *
- * Protobuf type {@code tensorflow.profiler.Resource}
- */
-public  final class Resource extends
-    com.google.protobuf.GeneratedMessageV3 implements
-    // @@protoc_insertion_point(message_implements:tensorflow.profiler.Resource)
-    ResourceOrBuilder {
-private static final long serialVersionUID = 0L;
-  // Use Resource.newBuilder() to construct.
-  private Resource(com.google.protobuf.GeneratedMessageV3.Builder<?> builder) {
-    super(builder);
-  }
-  private Resource() {
-    name_ = "";
-  }
-
-  @java.lang.Override
-  @SuppressWarnings({"unused"})
-  protected java.lang.Object newInstance(
-      UnusedPrivateParameter unused) {
-    return new Resource();
-  }
-
-  @java.lang.Override
-  public final com.google.protobuf.UnknownFieldSet
-  getUnknownFields() {
-    return this.unknownFields;
-  }
-  private Resource(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    this();
-    if (extensionRegistry == null) {
-      throw new java.lang.NullPointerException();
-    }
-    com.google.protobuf.UnknownFieldSet.Builder unknownFields =
-        com.google.protobuf.UnknownFieldSet.newBuilder();
-    try {
-      boolean done = false;
-      while (!done) {
-        int tag = input.readTag();
-        switch (tag) {
-          case 0:
-            done = true;
-            break;
-          case 10: {
-            java.lang.String s = input.readStringRequireUtf8();
-
-            name_ = s;
-            break;
-          }
-          case 16: {
-
-            resourceId_ = input.readUInt32();
-            break;
-          }
-          default: {
-            if (!parseUnknownField(
-                input, unknownFields, extensionRegistry, tag)) {
-              done = true;
-            }
-            break;
-          }
-        }
-      }
-    } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-      throw e.setUnfinishedMessage(this);
-    } catch (java.io.IOException e) {
-      throw new com.google.protobuf.InvalidProtocolBufferException(
-          e).setUnfinishedMessage(this);
-    } finally {
-      this.unknownFields = unknownFields.build();
-      makeExtensionsImmutable();
-    }
-  }
-  public static final com.google.protobuf.Descriptors.Descriptor
-      getDescriptor() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Resource_descriptor;
-  }
-
-  @java.lang.Override
-  protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internalGetFieldAccessorTable() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Resource_fieldAccessorTable
-        .ensureFieldAccessorsInitialized(
-            org.tensorflow.proto.framework.Resource.class, org.tensorflow.proto.framework.Resource.Builder.class);
-  }
-
-  public static final int NAME_FIELD_NUMBER = 1;
-  private volatile java.lang.Object name_;
-  /**
-   * <pre>
-   * The name of the resource.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  public java.lang.String getName() {
-    java.lang.Object ref = name_;
-    if (ref instanceof java.lang.String) {
-      return (java.lang.String) ref;
-    } else {
-      com.google.protobuf.ByteString bs = 
-          (com.google.protobuf.ByteString) ref;
-      java.lang.String s = bs.toStringUtf8();
-      name_ = s;
-      return s;
-    }
-  }
-  /**
-   * <pre>
-   * The name of the resource.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  public com.google.protobuf.ByteString
-      getNameBytes() {
-    java.lang.Object ref = name_;
-    if (ref instanceof java.lang.String) {
-      com.google.protobuf.ByteString b = 
-          com.google.protobuf.ByteString.copyFromUtf8(
-              (java.lang.String) ref);
-      name_ = b;
-      return b;
-    } else {
-      return (com.google.protobuf.ByteString) ref;
-    }
-  }
-
-  public static final int RESOURCE_ID_FIELD_NUMBER = 2;
-  private int resourceId_;
-  /**
-   * <pre>
-   * The id of the resource. Unique within a device.
-   * </pre>
-   *
-   * <code>uint32 resource_id = 2;</code>
-   */
-  public int getResourceId() {
-    return resourceId_;
-  }
-
-  private byte memoizedIsInitialized = -1;
-  @java.lang.Override
-  public final boolean isInitialized() {
-    byte isInitialized = memoizedIsInitialized;
-    if (isInitialized == 1) return true;
-    if (isInitialized == 0) return false;
-
-    memoizedIsInitialized = 1;
-    return true;
-  }
-
-  @java.lang.Override
-  public void writeTo(com.google.protobuf.CodedOutputStream output)
-                      throws java.io.IOException {
-    if (!getNameBytes().isEmpty()) {
-      com.google.protobuf.GeneratedMessageV3.writeString(output, 1, name_);
-    }
-    if (resourceId_ != 0) {
-      output.writeUInt32(2, resourceId_);
-    }
-    unknownFields.writeTo(output);
-  }
-
-  @java.lang.Override
-  public int getSerializedSize() {
-    int size = memoizedSize;
-    if (size != -1) return size;
-
-    size = 0;
-    if (!getNameBytes().isEmpty()) {
-      size += com.google.protobuf.GeneratedMessageV3.computeStringSize(1, name_);
-    }
-    if (resourceId_ != 0) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeUInt32Size(2, resourceId_);
-    }
-    size += unknownFields.getSerializedSize();
-    memoizedSize = size;
-    return size;
-  }
-
-  @java.lang.Override
-  public boolean equals(final java.lang.Object obj) {
-    if (obj == this) {
-     return true;
-    }
-    if (!(obj instanceof org.tensorflow.proto.framework.Resource)) {
-      return super.equals(obj);
-    }
-    org.tensorflow.proto.framework.Resource other = (org.tensorflow.proto.framework.Resource) obj;
-
-    if (!getName()
-        .equals(other.getName())) return false;
-    if (getResourceId()
-        != other.getResourceId()) return false;
-    if (!unknownFields.equals(other.unknownFields)) return false;
-    return true;
-  }
-
-  @java.lang.Override
-  public int hashCode() {
-    if (memoizedHashCode != 0) {
-      return memoizedHashCode;
-    }
-    int hash = 41;
-    hash = (19 * hash) + getDescriptor().hashCode();
-    hash = (37 * hash) + NAME_FIELD_NUMBER;
-    hash = (53 * hash) + getName().hashCode();
-    hash = (37 * hash) + RESOURCE_ID_FIELD_NUMBER;
-    hash = (53 * hash) + getResourceId();
-    hash = (29 * hash) + unknownFields.hashCode();
-    memoizedHashCode = hash;
-    return hash;
-  }
-
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      java.nio.ByteBuffer data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      java.nio.ByteBuffer data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      com.google.protobuf.ByteString data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      com.google.protobuf.ByteString data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(byte[] data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      byte[] data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Resource parseDelimitedFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Resource parseDelimitedFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      com.google.protobuf.CodedInputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Resource parseFrom(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-
-  @java.lang.Override
-  public Builder newBuilderForType() { return newBuilder(); }
-  public static Builder newBuilder() {
-    return DEFAULT_INSTANCE.toBuilder();
-  }
-  public static Builder newBuilder(org.tensorflow.proto.framework.Resource prototype) {
-    return DEFAULT_INSTANCE.toBuilder().mergeFrom(prototype);
-  }
-  @java.lang.Override
-  public Builder toBuilder() {
-    return this == DEFAULT_INSTANCE
-        ? new Builder() : new Builder().mergeFrom(this);
-  }
-
-  @java.lang.Override
-  protected Builder newBuilderForType(
-      com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-    Builder builder = new Builder(parent);
-    return builder;
-  }
-  /**
-   * <pre>
-   * A 'resource' generally is a specific computation component on a device. These
-   * can range from threads on CPUs to specific arithmetic units on hardware
-   * devices.
-   * </pre>
-   *
-   * Protobuf type {@code tensorflow.profiler.Resource}
-   */
-  public static final class Builder extends
-      com.google.protobuf.GeneratedMessageV3.Builder<Builder> implements
-      // @@protoc_insertion_point(builder_implements:tensorflow.profiler.Resource)
-      org.tensorflow.proto.framework.ResourceOrBuilder {
-    public static final com.google.protobuf.Descriptors.Descriptor
-        getDescriptor() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Resource_descriptor;
-    }
-
-    @java.lang.Override
-    protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-        internalGetFieldAccessorTable() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Resource_fieldAccessorTable
-          .ensureFieldAccessorsInitialized(
-              org.tensorflow.proto.framework.Resource.class, org.tensorflow.proto.framework.Resource.Builder.class);
-    }
-
-    // Construct using org.tensorflow.proto.framework.Resource.newBuilder()
-    private Builder() {
-      maybeForceBuilderInitialization();
-    }
-
-    private Builder(
-        com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-      super(parent);
-      maybeForceBuilderInitialization();
-    }
-    private void maybeForceBuilderInitialization() {
-      if (com.google.protobuf.GeneratedMessageV3
-              .alwaysUseFieldBuilders) {
-      }
-    }
-    @java.lang.Override
-    public Builder clear() {
-      super.clear();
-      name_ = "";
-
-      resourceId_ = 0;
-
-      return this;
-    }
-
-    @java.lang.Override
-    public com.google.protobuf.Descriptors.Descriptor
-        getDescriptorForType() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Resource_descriptor;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Resource getDefaultInstanceForType() {
-      return org.tensorflow.proto.framework.Resource.getDefaultInstance();
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Resource build() {
-      org.tensorflow.proto.framework.Resource result = buildPartial();
-      if (!result.isInitialized()) {
-        throw newUninitializedMessageException(result);
-      }
-      return result;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Resource buildPartial() {
-      org.tensorflow.proto.framework.Resource result = new org.tensorflow.proto.framework.Resource(this);
-      result.name_ = name_;
-      result.resourceId_ = resourceId_;
-      onBuilt();
-      return result;
-    }
-
-    @java.lang.Override
-    public Builder clone() {
-      return super.clone();
-    }
-    @java.lang.Override
-    public Builder setField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.setField(field, value);
-    }
-    @java.lang.Override
-    public Builder clearField(
-        com.google.protobuf.Descriptors.FieldDescriptor field) {
-      return super.clearField(field);
-    }
-    @java.lang.Override
-    public Builder clearOneof(
-        com.google.protobuf.Descriptors.OneofDescriptor oneof) {
-      return super.clearOneof(oneof);
-    }
-    @java.lang.Override
-    public Builder setRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        int index, java.lang.Object value) {
-      return super.setRepeatedField(field, index, value);
-    }
-    @java.lang.Override
-    public Builder addRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.addRepeatedField(field, value);
-    }
-    @java.lang.Override
-    public Builder mergeFrom(com.google.protobuf.Message other) {
-      if (other instanceof org.tensorflow.proto.framework.Resource) {
-        return mergeFrom((org.tensorflow.proto.framework.Resource)other);
-      } else {
-        super.mergeFrom(other);
-        return this;
-      }
-    }
-
-    public Builder mergeFrom(org.tensorflow.proto.framework.Resource other) {
-      if (other == org.tensorflow.proto.framework.Resource.getDefaultInstance()) return this;
-      if (!other.getName().isEmpty()) {
-        name_ = other.name_;
-        onChanged();
-      }
-      if (other.getResourceId() != 0) {
-        setResourceId(other.getResourceId());
-      }
-      this.mergeUnknownFields(other.unknownFields);
-      onChanged();
-      return this;
-    }
-
-    @java.lang.Override
-    public final boolean isInitialized() {
-      return true;
-    }
-
-    @java.lang.Override
-    public Builder mergeFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws java.io.IOException {
-      org.tensorflow.proto.framework.Resource parsedMessage = null;
-      try {
-        parsedMessage = PARSER.parsePartialFrom(input, extensionRegistry);
-      } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-        parsedMessage = (org.tensorflow.proto.framework.Resource) e.getUnfinishedMessage();
-        throw e.unwrapIOException();
-      } finally {
-        if (parsedMessage != null) {
-          mergeFrom(parsedMessage);
-        }
-      }
-      return this;
-    }
-
-    private java.lang.Object name_ = "";
-    /**
-     * <pre>
-     * The name of the resource.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public java.lang.String getName() {
-      java.lang.Object ref = name_;
-      if (!(ref instanceof java.lang.String)) {
-        com.google.protobuf.ByteString bs =
-            (com.google.protobuf.ByteString) ref;
-        java.lang.String s = bs.toStringUtf8();
-        name_ = s;
-        return s;
-      } else {
-        return (java.lang.String) ref;
-      }
-    }
-    /**
-     * <pre>
-     * The name of the resource.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public com.google.protobuf.ByteString
-        getNameBytes() {
-      java.lang.Object ref = name_;
-      if (ref instanceof String) {
-        com.google.protobuf.ByteString b = 
-            com.google.protobuf.ByteString.copyFromUtf8(
-                (java.lang.String) ref);
-        name_ = b;
-        return b;
-      } else {
-        return (com.google.protobuf.ByteString) ref;
-      }
-    }
-    /**
-     * <pre>
-     * The name of the resource.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public Builder setName(
-        java.lang.String value) {
-      if (value == null) {
-    throw new NullPointerException();
-  }
-  
-      name_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The name of the resource.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public Builder clearName() {
-      
-      name_ = getDefaultInstance().getName();
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The name of the resource.
-     * </pre>
-     *
-     * <code>string name = 1;</code>
-     */
-    public Builder setNameBytes(
-        com.google.protobuf.ByteString value) {
-      if (value == null) {
-    throw new NullPointerException();
-  }
-  checkByteStringIsUtf8(value);
-      
-      name_ = value;
-      onChanged();
-      return this;
-    }
-
-    private int resourceId_ ;
-    /**
-     * <pre>
-     * The id of the resource. Unique within a device.
-     * </pre>
-     *
-     * <code>uint32 resource_id = 2;</code>
-     */
-    public int getResourceId() {
-      return resourceId_;
-    }
-    /**
-     * <pre>
-     * The id of the resource. Unique within a device.
-     * </pre>
-     *
-     * <code>uint32 resource_id = 2;</code>
-     */
-    public Builder setResourceId(int value) {
-      
-      resourceId_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The id of the resource. Unique within a device.
-     * </pre>
-     *
-     * <code>uint32 resource_id = 2;</code>
-     */
-    public Builder clearResourceId() {
-      
-      resourceId_ = 0;
-      onChanged();
-      return this;
-    }
-    @java.lang.Override
-    public final Builder setUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.setUnknownFields(unknownFields);
-    }
-
-    @java.lang.Override
-    public final Builder mergeUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.mergeUnknownFields(unknownFields);
-    }
-
-
-    // @@protoc_insertion_point(builder_scope:tensorflow.profiler.Resource)
-  }
-
-  // @@protoc_insertion_point(class_scope:tensorflow.profiler.Resource)
-  private static final org.tensorflow.proto.framework.Resource DEFAULT_INSTANCE;
-  static {
-    DEFAULT_INSTANCE = new org.tensorflow.proto.framework.Resource();
-  }
-
-  public static org.tensorflow.proto.framework.Resource getDefaultInstance() {
-    return DEFAULT_INSTANCE;
-  }
-
-  private static final com.google.protobuf.Parser<Resource>
-      PARSER = new com.google.protobuf.AbstractParser<Resource>() {
-    @java.lang.Override
-    public Resource parsePartialFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws com.google.protobuf.InvalidProtocolBufferException {
-      return new Resource(input, extensionRegistry);
-    }
-  };
-
-  public static com.google.protobuf.Parser<Resource> parser() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public com.google.protobuf.Parser<Resource> getParserForType() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public org.tensorflow.proto.framework.Resource getDefaultInstanceForType() {
-    return DEFAULT_INSTANCE;
-  }
-
-}
-
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/ResourceOrBuilder.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/ResourceOrBuilder.java
deleted file mode 100644
index f6bc0c0d190..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/ResourceOrBuilder.java
+++ /dev/null
@@ -1,36 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-public interface ResourceOrBuilder extends
-    // @@protoc_insertion_point(interface_extends:tensorflow.profiler.Resource)
-    com.google.protobuf.MessageOrBuilder {
-
-  /**
-   * <pre>
-   * The name of the resource.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  java.lang.String getName();
-  /**
-   * <pre>
-   * The name of the resource.
-   * </pre>
-   *
-   * <code>string name = 1;</code>
-   */
-  com.google.protobuf.ByteString
-      getNameBytes();
-
-  /**
-   * <pre>
-   * The id of the resource. Unique within a device.
-   * </pre>
-   *
-   * <code>uint32 resource_id = 2;</code>
-   */
-  int getResourceId();
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Trace.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Trace.java
deleted file mode 100644
index 09d97221dde..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/Trace.java
+++ /dev/null
@@ -1,1193 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-/**
- * <pre>
- * A 'Trace' contains metadata for the individual traces of a system.
- * </pre>
- *
- * Protobuf type {@code tensorflow.profiler.Trace}
- */
-public  final class Trace extends
-    com.google.protobuf.GeneratedMessageV3 implements
-    // @@protoc_insertion_point(message_implements:tensorflow.profiler.Trace)
-    TraceOrBuilder {
-private static final long serialVersionUID = 0L;
-  // Use Trace.newBuilder() to construct.
-  private Trace(com.google.protobuf.GeneratedMessageV3.Builder<?> builder) {
-    super(builder);
-  }
-  private Trace() {
-    traceEvents_ = java.util.Collections.emptyList();
-  }
-
-  @java.lang.Override
-  @SuppressWarnings({"unused"})
-  protected java.lang.Object newInstance(
-      UnusedPrivateParameter unused) {
-    return new Trace();
-  }
-
-  @java.lang.Override
-  public final com.google.protobuf.UnknownFieldSet
-  getUnknownFields() {
-    return this.unknownFields;
-  }
-  private Trace(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    this();
-    if (extensionRegistry == null) {
-      throw new java.lang.NullPointerException();
-    }
-    int mutable_bitField0_ = 0;
-    com.google.protobuf.UnknownFieldSet.Builder unknownFields =
-        com.google.protobuf.UnknownFieldSet.newBuilder();
-    try {
-      boolean done = false;
-      while (!done) {
-        int tag = input.readTag();
-        switch (tag) {
-          case 0:
-            done = true;
-            break;
-          case 10: {
-            if (!((mutable_bitField0_ & 0x00000001) != 0)) {
-              devices_ = com.google.protobuf.MapField.newMapField(
-                  DevicesDefaultEntryHolder.defaultEntry);
-              mutable_bitField0_ |= 0x00000001;
-            }
-            com.google.protobuf.MapEntry<java.lang.Integer, org.tensorflow.proto.framework.Device>
-            devices__ = input.readMessage(
-                DevicesDefaultEntryHolder.defaultEntry.getParserForType(), extensionRegistry);
-            devices_.getMutableMap().put(
-                devices__.getKey(), devices__.getValue());
-            break;
-          }
-          case 34: {
-            if (!((mutable_bitField0_ & 0x00000002) != 0)) {
-              traceEvents_ = new java.util.ArrayList<org.tensorflow.proto.framework.TraceEvent>();
-              mutable_bitField0_ |= 0x00000002;
-            }
-            traceEvents_.add(
-                input.readMessage(org.tensorflow.proto.framework.TraceEvent.parser(), extensionRegistry));
-            break;
-          }
-          default: {
-            if (!parseUnknownField(
-                input, unknownFields, extensionRegistry, tag)) {
-              done = true;
-            }
-            break;
-          }
-        }
-      }
-    } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-      throw e.setUnfinishedMessage(this);
-    } catch (java.io.IOException e) {
-      throw new com.google.protobuf.InvalidProtocolBufferException(
-          e).setUnfinishedMessage(this);
-    } finally {
-      if (((mutable_bitField0_ & 0x00000002) != 0)) {
-        traceEvents_ = java.util.Collections.unmodifiableList(traceEvents_);
-      }
-      this.unknownFields = unknownFields.build();
-      makeExtensionsImmutable();
-    }
-  }
-  public static final com.google.protobuf.Descriptors.Descriptor
-      getDescriptor() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Trace_descriptor;
-  }
-
-  @SuppressWarnings({"rawtypes"})
-  @java.lang.Override
-  protected com.google.protobuf.MapField internalGetMapField(
-      int number) {
-    switch (number) {
-      case 1:
-        return internalGetDevices();
-      default:
-        throw new RuntimeException(
-            "Invalid map field number: " + number);
-    }
-  }
-  @java.lang.Override
-  protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internalGetFieldAccessorTable() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Trace_fieldAccessorTable
-        .ensureFieldAccessorsInitialized(
-            org.tensorflow.proto.framework.Trace.class, org.tensorflow.proto.framework.Trace.Builder.class);
-  }
-
-  public static final int DEVICES_FIELD_NUMBER = 1;
-  private static final class DevicesDefaultEntryHolder {
-    static final com.google.protobuf.MapEntry<
-        java.lang.Integer, org.tensorflow.proto.framework.Device> defaultEntry =
-            com.google.protobuf.MapEntry
-            .<java.lang.Integer, org.tensorflow.proto.framework.Device>newDefaultInstance(
-                org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Trace_DevicesEntry_descriptor, 
-                com.google.protobuf.WireFormat.FieldType.UINT32,
-                0,
-                com.google.protobuf.WireFormat.FieldType.MESSAGE,
-                org.tensorflow.proto.framework.Device.getDefaultInstance());
-  }
-  private com.google.protobuf.MapField<
-      java.lang.Integer, org.tensorflow.proto.framework.Device> devices_;
-  private com.google.protobuf.MapField<java.lang.Integer, org.tensorflow.proto.framework.Device>
-  internalGetDevices() {
-    if (devices_ == null) {
-      return com.google.protobuf.MapField.emptyMapField(
-          DevicesDefaultEntryHolder.defaultEntry);
-    }
-    return devices_;
-  }
-
-  public int getDevicesCount() {
-    return internalGetDevices().getMap().size();
-  }
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-
-  public boolean containsDevices(
-      int key) {
-    
-    return internalGetDevices().getMap().containsKey(key);
-  }
-  /**
-   * Use {@link #getDevicesMap()} instead.
-   */
-  @java.lang.Deprecated
-  public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> getDevices() {
-    return getDevicesMap();
-  }
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-
-  public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> getDevicesMap() {
-    return internalGetDevices().getMap();
-  }
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-
-  public org.tensorflow.proto.framework.Device getDevicesOrDefault(
-      int key,
-      org.tensorflow.proto.framework.Device defaultValue) {
-    
-    java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> map =
-        internalGetDevices().getMap();
-    return map.containsKey(key) ? map.get(key) : defaultValue;
-  }
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-
-  public org.tensorflow.proto.framework.Device getDevicesOrThrow(
-      int key) {
-    
-    java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> map =
-        internalGetDevices().getMap();
-    if (!map.containsKey(key)) {
-      throw new java.lang.IllegalArgumentException();
-    }
-    return map.get(key);
-  }
-
-  public static final int TRACE_EVENTS_FIELD_NUMBER = 4;
-  private java.util.List<org.tensorflow.proto.framework.TraceEvent> traceEvents_;
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  public java.util.List<org.tensorflow.proto.framework.TraceEvent> getTraceEventsList() {
-    return traceEvents_;
-  }
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  public java.util.List<? extends org.tensorflow.proto.framework.TraceEventOrBuilder> 
-      getTraceEventsOrBuilderList() {
-    return traceEvents_;
-  }
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  public int getTraceEventsCount() {
-    return traceEvents_.size();
-  }
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  public org.tensorflow.proto.framework.TraceEvent getTraceEvents(int index) {
-    return traceEvents_.get(index);
-  }
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  public org.tensorflow.proto.framework.TraceEventOrBuilder getTraceEventsOrBuilder(
-      int index) {
-    return traceEvents_.get(index);
-  }
-
-  private byte memoizedIsInitialized = -1;
-  @java.lang.Override
-  public final boolean isInitialized() {
-    byte isInitialized = memoizedIsInitialized;
-    if (isInitialized == 1) return true;
-    if (isInitialized == 0) return false;
-
-    memoizedIsInitialized = 1;
-    return true;
-  }
-
-  @java.lang.Override
-  public void writeTo(com.google.protobuf.CodedOutputStream output)
-                      throws java.io.IOException {
-    com.google.protobuf.GeneratedMessageV3
-      .serializeIntegerMapTo(
-        output,
-        internalGetDevices(),
-        DevicesDefaultEntryHolder.defaultEntry,
-        1);
-    for (int i = 0; i < traceEvents_.size(); i++) {
-      output.writeMessage(4, traceEvents_.get(i));
-    }
-    unknownFields.writeTo(output);
-  }
-
-  @java.lang.Override
-  public int getSerializedSize() {
-    int size = memoizedSize;
-    if (size != -1) return size;
-
-    size = 0;
-    for (java.util.Map.Entry<java.lang.Integer, org.tensorflow.proto.framework.Device> entry
-         : internalGetDevices().getMap().entrySet()) {
-      com.google.protobuf.MapEntry<java.lang.Integer, org.tensorflow.proto.framework.Device>
-      devices__ = DevicesDefaultEntryHolder.defaultEntry.newBuilderForType()
-          .setKey(entry.getKey())
-          .setValue(entry.getValue())
-          .build();
-      size += com.google.protobuf.CodedOutputStream
-          .computeMessageSize(1, devices__);
-    }
-    for (int i = 0; i < traceEvents_.size(); i++) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeMessageSize(4, traceEvents_.get(i));
-    }
-    size += unknownFields.getSerializedSize();
-    memoizedSize = size;
-    return size;
-  }
-
-  @java.lang.Override
-  public boolean equals(final java.lang.Object obj) {
-    if (obj == this) {
-     return true;
-    }
-    if (!(obj instanceof org.tensorflow.proto.framework.Trace)) {
-      return super.equals(obj);
-    }
-    org.tensorflow.proto.framework.Trace other = (org.tensorflow.proto.framework.Trace) obj;
-
-    if (!internalGetDevices().equals(
-        other.internalGetDevices())) return false;
-    if (!getTraceEventsList()
-        .equals(other.getTraceEventsList())) return false;
-    if (!unknownFields.equals(other.unknownFields)) return false;
-    return true;
-  }
-
-  @java.lang.Override
-  public int hashCode() {
-    if (memoizedHashCode != 0) {
-      return memoizedHashCode;
-    }
-    int hash = 41;
-    hash = (19 * hash) + getDescriptor().hashCode();
-    if (!internalGetDevices().getMap().isEmpty()) {
-      hash = (37 * hash) + DEVICES_FIELD_NUMBER;
-      hash = (53 * hash) + internalGetDevices().hashCode();
-    }
-    if (getTraceEventsCount() > 0) {
-      hash = (37 * hash) + TRACE_EVENTS_FIELD_NUMBER;
-      hash = (53 * hash) + getTraceEventsList().hashCode();
-    }
-    hash = (29 * hash) + unknownFields.hashCode();
-    memoizedHashCode = hash;
-    return hash;
-  }
-
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      java.nio.ByteBuffer data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      java.nio.ByteBuffer data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      com.google.protobuf.ByteString data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      com.google.protobuf.ByteString data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(byte[] data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      byte[] data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Trace parseDelimitedFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Trace parseDelimitedFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      com.google.protobuf.CodedInputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.Trace parseFrom(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-
-  @java.lang.Override
-  public Builder newBuilderForType() { return newBuilder(); }
-  public static Builder newBuilder() {
-    return DEFAULT_INSTANCE.toBuilder();
-  }
-  public static Builder newBuilder(org.tensorflow.proto.framework.Trace prototype) {
-    return DEFAULT_INSTANCE.toBuilder().mergeFrom(prototype);
-  }
-  @java.lang.Override
-  public Builder toBuilder() {
-    return this == DEFAULT_INSTANCE
-        ? new Builder() : new Builder().mergeFrom(this);
-  }
-
-  @java.lang.Override
-  protected Builder newBuilderForType(
-      com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-    Builder builder = new Builder(parent);
-    return builder;
-  }
-  /**
-   * <pre>
-   * A 'Trace' contains metadata for the individual traces of a system.
-   * </pre>
-   *
-   * Protobuf type {@code tensorflow.profiler.Trace}
-   */
-  public static final class Builder extends
-      com.google.protobuf.GeneratedMessageV3.Builder<Builder> implements
-      // @@protoc_insertion_point(builder_implements:tensorflow.profiler.Trace)
-      org.tensorflow.proto.framework.TraceOrBuilder {
-    public static final com.google.protobuf.Descriptors.Descriptor
-        getDescriptor() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Trace_descriptor;
-    }
-
-    @SuppressWarnings({"rawtypes"})
-    protected com.google.protobuf.MapField internalGetMapField(
-        int number) {
-      switch (number) {
-        case 1:
-          return internalGetDevices();
-        default:
-          throw new RuntimeException(
-              "Invalid map field number: " + number);
-      }
-    }
-    @SuppressWarnings({"rawtypes"})
-    protected com.google.protobuf.MapField internalGetMutableMapField(
-        int number) {
-      switch (number) {
-        case 1:
-          return internalGetMutableDevices();
-        default:
-          throw new RuntimeException(
-              "Invalid map field number: " + number);
-      }
-    }
-    @java.lang.Override
-    protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-        internalGetFieldAccessorTable() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Trace_fieldAccessorTable
-          .ensureFieldAccessorsInitialized(
-              org.tensorflow.proto.framework.Trace.class, org.tensorflow.proto.framework.Trace.Builder.class);
-    }
-
-    // Construct using org.tensorflow.proto.framework.Trace.newBuilder()
-    private Builder() {
-      maybeForceBuilderInitialization();
-    }
-
-    private Builder(
-        com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-      super(parent);
-      maybeForceBuilderInitialization();
-    }
-    private void maybeForceBuilderInitialization() {
-      if (com.google.protobuf.GeneratedMessageV3
-              .alwaysUseFieldBuilders) {
-        getTraceEventsFieldBuilder();
-      }
-    }
-    @java.lang.Override
-    public Builder clear() {
-      super.clear();
-      internalGetMutableDevices().clear();
-      if (traceEventsBuilder_ == null) {
-        traceEvents_ = java.util.Collections.emptyList();
-        bitField0_ = (bitField0_ & ~0x00000002);
-      } else {
-        traceEventsBuilder_.clear();
-      }
-      return this;
-    }
-
-    @java.lang.Override
-    public com.google.protobuf.Descriptors.Descriptor
-        getDescriptorForType() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_Trace_descriptor;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Trace getDefaultInstanceForType() {
-      return org.tensorflow.proto.framework.Trace.getDefaultInstance();
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Trace build() {
-      org.tensorflow.proto.framework.Trace result = buildPartial();
-      if (!result.isInitialized()) {
-        throw newUninitializedMessageException(result);
-      }
-      return result;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.Trace buildPartial() {
-      org.tensorflow.proto.framework.Trace result = new org.tensorflow.proto.framework.Trace(this);
-      int from_bitField0_ = bitField0_;
-      result.devices_ = internalGetDevices();
-      result.devices_.makeImmutable();
-      if (traceEventsBuilder_ == null) {
-        if (((bitField0_ & 0x00000002) != 0)) {
-          traceEvents_ = java.util.Collections.unmodifiableList(traceEvents_);
-          bitField0_ = (bitField0_ & ~0x00000002);
-        }
-        result.traceEvents_ = traceEvents_;
-      } else {
-        result.traceEvents_ = traceEventsBuilder_.build();
-      }
-      onBuilt();
-      return result;
-    }
-
-    @java.lang.Override
-    public Builder clone() {
-      return super.clone();
-    }
-    @java.lang.Override
-    public Builder setField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.setField(field, value);
-    }
-    @java.lang.Override
-    public Builder clearField(
-        com.google.protobuf.Descriptors.FieldDescriptor field) {
-      return super.clearField(field);
-    }
-    @java.lang.Override
-    public Builder clearOneof(
-        com.google.protobuf.Descriptors.OneofDescriptor oneof) {
-      return super.clearOneof(oneof);
-    }
-    @java.lang.Override
-    public Builder setRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        int index, java.lang.Object value) {
-      return super.setRepeatedField(field, index, value);
-    }
-    @java.lang.Override
-    public Builder addRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.addRepeatedField(field, value);
-    }
-    @java.lang.Override
-    public Builder mergeFrom(com.google.protobuf.Message other) {
-      if (other instanceof org.tensorflow.proto.framework.Trace) {
-        return mergeFrom((org.tensorflow.proto.framework.Trace)other);
-      } else {
-        super.mergeFrom(other);
-        return this;
-      }
-    }
-
-    public Builder mergeFrom(org.tensorflow.proto.framework.Trace other) {
-      if (other == org.tensorflow.proto.framework.Trace.getDefaultInstance()) return this;
-      internalGetMutableDevices().mergeFrom(
-          other.internalGetDevices());
-      if (traceEventsBuilder_ == null) {
-        if (!other.traceEvents_.isEmpty()) {
-          if (traceEvents_.isEmpty()) {
-            traceEvents_ = other.traceEvents_;
-            bitField0_ = (bitField0_ & ~0x00000002);
-          } else {
-            ensureTraceEventsIsMutable();
-            traceEvents_.addAll(other.traceEvents_);
-          }
-          onChanged();
-        }
-      } else {
-        if (!other.traceEvents_.isEmpty()) {
-          if (traceEventsBuilder_.isEmpty()) {
-            traceEventsBuilder_.dispose();
-            traceEventsBuilder_ = null;
-            traceEvents_ = other.traceEvents_;
-            bitField0_ = (bitField0_ & ~0x00000002);
-            traceEventsBuilder_ = 
-              com.google.protobuf.GeneratedMessageV3.alwaysUseFieldBuilders ?
-                 getTraceEventsFieldBuilder() : null;
-          } else {
-            traceEventsBuilder_.addAllMessages(other.traceEvents_);
-          }
-        }
-      }
-      this.mergeUnknownFields(other.unknownFields);
-      onChanged();
-      return this;
-    }
-
-    @java.lang.Override
-    public final boolean isInitialized() {
-      return true;
-    }
-
-    @java.lang.Override
-    public Builder mergeFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws java.io.IOException {
-      org.tensorflow.proto.framework.Trace parsedMessage = null;
-      try {
-        parsedMessage = PARSER.parsePartialFrom(input, extensionRegistry);
-      } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-        parsedMessage = (org.tensorflow.proto.framework.Trace) e.getUnfinishedMessage();
-        throw e.unwrapIOException();
-      } finally {
-        if (parsedMessage != null) {
-          mergeFrom(parsedMessage);
-        }
-      }
-      return this;
-    }
-    private int bitField0_;
-
-    private com.google.protobuf.MapField<
-        java.lang.Integer, org.tensorflow.proto.framework.Device> devices_;
-    private com.google.protobuf.MapField<java.lang.Integer, org.tensorflow.proto.framework.Device>
-    internalGetDevices() {
-      if (devices_ == null) {
-        return com.google.protobuf.MapField.emptyMapField(
-            DevicesDefaultEntryHolder.defaultEntry);
-      }
-      return devices_;
-    }
-    private com.google.protobuf.MapField<java.lang.Integer, org.tensorflow.proto.framework.Device>
-    internalGetMutableDevices() {
-      onChanged();;
-      if (devices_ == null) {
-        devices_ = com.google.protobuf.MapField.newMapField(
-            DevicesDefaultEntryHolder.defaultEntry);
-      }
-      if (!devices_.isMutable()) {
-        devices_ = devices_.copy();
-      }
-      return devices_;
-    }
-
-    public int getDevicesCount() {
-      return internalGetDevices().getMap().size();
-    }
-    /**
-     * <pre>
-     * The devices that this trace has information about. Maps from device_id to
-     * more data about the specific device.
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-     */
-
-    public boolean containsDevices(
-        int key) {
-      
-      return internalGetDevices().getMap().containsKey(key);
-    }
-    /**
-     * Use {@link #getDevicesMap()} instead.
-     */
-    @java.lang.Deprecated
-    public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> getDevices() {
-      return getDevicesMap();
-    }
-    /**
-     * <pre>
-     * The devices that this trace has information about. Maps from device_id to
-     * more data about the specific device.
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-     */
-
-    public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> getDevicesMap() {
-      return internalGetDevices().getMap();
-    }
-    /**
-     * <pre>
-     * The devices that this trace has information about. Maps from device_id to
-     * more data about the specific device.
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-     */
-
-    public org.tensorflow.proto.framework.Device getDevicesOrDefault(
-        int key,
-        org.tensorflow.proto.framework.Device defaultValue) {
-      
-      java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> map =
-          internalGetDevices().getMap();
-      return map.containsKey(key) ? map.get(key) : defaultValue;
-    }
-    /**
-     * <pre>
-     * The devices that this trace has information about. Maps from device_id to
-     * more data about the specific device.
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-     */
-
-    public org.tensorflow.proto.framework.Device getDevicesOrThrow(
-        int key) {
-      
-      java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> map =
-          internalGetDevices().getMap();
-      if (!map.containsKey(key)) {
-        throw new java.lang.IllegalArgumentException();
-      }
-      return map.get(key);
-    }
-
-    public Builder clearDevices() {
-      internalGetMutableDevices().getMutableMap()
-          .clear();
-      return this;
-    }
-    /**
-     * <pre>
-     * The devices that this trace has information about. Maps from device_id to
-     * more data about the specific device.
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-     */
-
-    public Builder removeDevices(
-        int key) {
-      
-      internalGetMutableDevices().getMutableMap()
-          .remove(key);
-      return this;
-    }
-    /**
-     * Use alternate mutation accessors instead.
-     */
-    @java.lang.Deprecated
-    public java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device>
-    getMutableDevices() {
-      return internalGetMutableDevices().getMutableMap();
-    }
-    /**
-     * <pre>
-     * The devices that this trace has information about. Maps from device_id to
-     * more data about the specific device.
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-     */
-    public Builder putDevices(
-        int key,
-        org.tensorflow.proto.framework.Device value) {
-      
-      if (value == null) { throw new java.lang.NullPointerException(); }
-      internalGetMutableDevices().getMutableMap()
-          .put(key, value);
-      return this;
-    }
-    /**
-     * <pre>
-     * The devices that this trace has information about. Maps from device_id to
-     * more data about the specific device.
-     * </pre>
-     *
-     * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-     */
-
-    public Builder putAllDevices(
-        java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device> values) {
-      internalGetMutableDevices().getMutableMap()
-          .putAll(values);
-      return this;
-    }
-
-    private java.util.List<org.tensorflow.proto.framework.TraceEvent> traceEvents_ =
-      java.util.Collections.emptyList();
-    private void ensureTraceEventsIsMutable() {
-      if (!((bitField0_ & 0x00000002) != 0)) {
-        traceEvents_ = new java.util.ArrayList<org.tensorflow.proto.framework.TraceEvent>(traceEvents_);
-        bitField0_ |= 0x00000002;
-       }
-    }
-
-    private com.google.protobuf.RepeatedFieldBuilderV3<
-        org.tensorflow.proto.framework.TraceEvent, org.tensorflow.proto.framework.TraceEvent.Builder, org.tensorflow.proto.framework.TraceEventOrBuilder> traceEventsBuilder_;
-
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public java.util.List<org.tensorflow.proto.framework.TraceEvent> getTraceEventsList() {
-      if (traceEventsBuilder_ == null) {
-        return java.util.Collections.unmodifiableList(traceEvents_);
-      } else {
-        return traceEventsBuilder_.getMessageList();
-      }
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public int getTraceEventsCount() {
-      if (traceEventsBuilder_ == null) {
-        return traceEvents_.size();
-      } else {
-        return traceEventsBuilder_.getCount();
-      }
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public org.tensorflow.proto.framework.TraceEvent getTraceEvents(int index) {
-      if (traceEventsBuilder_ == null) {
-        return traceEvents_.get(index);
-      } else {
-        return traceEventsBuilder_.getMessage(index);
-      }
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder setTraceEvents(
-        int index, org.tensorflow.proto.framework.TraceEvent value) {
-      if (traceEventsBuilder_ == null) {
-        if (value == null) {
-          throw new NullPointerException();
-        }
-        ensureTraceEventsIsMutable();
-        traceEvents_.set(index, value);
-        onChanged();
-      } else {
-        traceEventsBuilder_.setMessage(index, value);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder setTraceEvents(
-        int index, org.tensorflow.proto.framework.TraceEvent.Builder builderForValue) {
-      if (traceEventsBuilder_ == null) {
-        ensureTraceEventsIsMutable();
-        traceEvents_.set(index, builderForValue.build());
-        onChanged();
-      } else {
-        traceEventsBuilder_.setMessage(index, builderForValue.build());
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder addTraceEvents(org.tensorflow.proto.framework.TraceEvent value) {
-      if (traceEventsBuilder_ == null) {
-        if (value == null) {
-          throw new NullPointerException();
-        }
-        ensureTraceEventsIsMutable();
-        traceEvents_.add(value);
-        onChanged();
-      } else {
-        traceEventsBuilder_.addMessage(value);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder addTraceEvents(
-        int index, org.tensorflow.proto.framework.TraceEvent value) {
-      if (traceEventsBuilder_ == null) {
-        if (value == null) {
-          throw new NullPointerException();
-        }
-        ensureTraceEventsIsMutable();
-        traceEvents_.add(index, value);
-        onChanged();
-      } else {
-        traceEventsBuilder_.addMessage(index, value);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder addTraceEvents(
-        org.tensorflow.proto.framework.TraceEvent.Builder builderForValue) {
-      if (traceEventsBuilder_ == null) {
-        ensureTraceEventsIsMutable();
-        traceEvents_.add(builderForValue.build());
-        onChanged();
-      } else {
-        traceEventsBuilder_.addMessage(builderForValue.build());
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder addTraceEvents(
-        int index, org.tensorflow.proto.framework.TraceEvent.Builder builderForValue) {
-      if (traceEventsBuilder_ == null) {
-        ensureTraceEventsIsMutable();
-        traceEvents_.add(index, builderForValue.build());
-        onChanged();
-      } else {
-        traceEventsBuilder_.addMessage(index, builderForValue.build());
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder addAllTraceEvents(
-        java.lang.Iterable<? extends org.tensorflow.proto.framework.TraceEvent> values) {
-      if (traceEventsBuilder_ == null) {
-        ensureTraceEventsIsMutable();
-        com.google.protobuf.AbstractMessageLite.Builder.addAll(
-            values, traceEvents_);
-        onChanged();
-      } else {
-        traceEventsBuilder_.addAllMessages(values);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder clearTraceEvents() {
-      if (traceEventsBuilder_ == null) {
-        traceEvents_ = java.util.Collections.emptyList();
-        bitField0_ = (bitField0_ & ~0x00000002);
-        onChanged();
-      } else {
-        traceEventsBuilder_.clear();
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public Builder removeTraceEvents(int index) {
-      if (traceEventsBuilder_ == null) {
-        ensureTraceEventsIsMutable();
-        traceEvents_.remove(index);
-        onChanged();
-      } else {
-        traceEventsBuilder_.remove(index);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public org.tensorflow.proto.framework.TraceEvent.Builder getTraceEventsBuilder(
-        int index) {
-      return getTraceEventsFieldBuilder().getBuilder(index);
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public org.tensorflow.proto.framework.TraceEventOrBuilder getTraceEventsOrBuilder(
-        int index) {
-      if (traceEventsBuilder_ == null) {
-        return traceEvents_.get(index);  } else {
-        return traceEventsBuilder_.getMessageOrBuilder(index);
-      }
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public java.util.List<? extends org.tensorflow.proto.framework.TraceEventOrBuilder> 
-         getTraceEventsOrBuilderList() {
-      if (traceEventsBuilder_ != null) {
-        return traceEventsBuilder_.getMessageOrBuilderList();
-      } else {
-        return java.util.Collections.unmodifiableList(traceEvents_);
-      }
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public org.tensorflow.proto.framework.TraceEvent.Builder addTraceEventsBuilder() {
-      return getTraceEventsFieldBuilder().addBuilder(
-          org.tensorflow.proto.framework.TraceEvent.getDefaultInstance());
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public org.tensorflow.proto.framework.TraceEvent.Builder addTraceEventsBuilder(
-        int index) {
-      return getTraceEventsFieldBuilder().addBuilder(
-          index, org.tensorflow.proto.framework.TraceEvent.getDefaultInstance());
-    }
-    /**
-     * <pre>
-     * All trace events capturing in the profiling period.
-     * </pre>
-     *
-     * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-     */
-    public java.util.List<org.tensorflow.proto.framework.TraceEvent.Builder> 
-         getTraceEventsBuilderList() {
-      return getTraceEventsFieldBuilder().getBuilderList();
-    }
-    private com.google.protobuf.RepeatedFieldBuilderV3<
-        org.tensorflow.proto.framework.TraceEvent, org.tensorflow.proto.framework.TraceEvent.Builder, org.tensorflow.proto.framework.TraceEventOrBuilder> 
-        getTraceEventsFieldBuilder() {
-      if (traceEventsBuilder_ == null) {
-        traceEventsBuilder_ = new com.google.protobuf.RepeatedFieldBuilderV3<
-            org.tensorflow.proto.framework.TraceEvent, org.tensorflow.proto.framework.TraceEvent.Builder, org.tensorflow.proto.framework.TraceEventOrBuilder>(
-                traceEvents_,
-                ((bitField0_ & 0x00000002) != 0),
-                getParentForChildren(),
-                isClean());
-        traceEvents_ = null;
-      }
-      return traceEventsBuilder_;
-    }
-    @java.lang.Override
-    public final Builder setUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.setUnknownFields(unknownFields);
-    }
-
-    @java.lang.Override
-    public final Builder mergeUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.mergeUnknownFields(unknownFields);
-    }
-
-
-    // @@protoc_insertion_point(builder_scope:tensorflow.profiler.Trace)
-  }
-
-  // @@protoc_insertion_point(class_scope:tensorflow.profiler.Trace)
-  private static final org.tensorflow.proto.framework.Trace DEFAULT_INSTANCE;
-  static {
-    DEFAULT_INSTANCE = new org.tensorflow.proto.framework.Trace();
-  }
-
-  public static org.tensorflow.proto.framework.Trace getDefaultInstance() {
-    return DEFAULT_INSTANCE;
-  }
-
-  private static final com.google.protobuf.Parser<Trace>
-      PARSER = new com.google.protobuf.AbstractParser<Trace>() {
-    @java.lang.Override
-    public Trace parsePartialFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws com.google.protobuf.InvalidProtocolBufferException {
-      return new Trace(input, extensionRegistry);
-    }
-  };
-
-  public static com.google.protobuf.Parser<Trace> parser() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public com.google.protobuf.Parser<Trace> getParserForType() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public org.tensorflow.proto.framework.Trace getDefaultInstanceForType() {
-    return DEFAULT_INSTANCE;
-  }
-
-}
-
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEvent.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEvent.java
deleted file mode 100644
index 8f34acfde34..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEvent.java
+++ /dev/null
@@ -1,1208 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-/**
- * Protobuf type {@code tensorflow.profiler.TraceEvent}
- */
-public  final class TraceEvent extends
-    com.google.protobuf.GeneratedMessageV3 implements
-    // @@protoc_insertion_point(message_implements:tensorflow.profiler.TraceEvent)
-    TraceEventOrBuilder {
-private static final long serialVersionUID = 0L;
-  // Use TraceEvent.newBuilder() to construct.
-  private TraceEvent(com.google.protobuf.GeneratedMessageV3.Builder<?> builder) {
-    super(builder);
-  }
-  private TraceEvent() {
-    name_ = "";
-  }
-
-  @java.lang.Override
-  @SuppressWarnings({"unused"})
-  protected java.lang.Object newInstance(
-      UnusedPrivateParameter unused) {
-    return new TraceEvent();
-  }
-
-  @java.lang.Override
-  public final com.google.protobuf.UnknownFieldSet
-  getUnknownFields() {
-    return this.unknownFields;
-  }
-  private TraceEvent(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    this();
-    if (extensionRegistry == null) {
-      throw new java.lang.NullPointerException();
-    }
-    int mutable_bitField0_ = 0;
-    com.google.protobuf.UnknownFieldSet.Builder unknownFields =
-        com.google.protobuf.UnknownFieldSet.newBuilder();
-    try {
-      boolean done = false;
-      while (!done) {
-        int tag = input.readTag();
-        switch (tag) {
-          case 0:
-            done = true;
-            break;
-          case 8: {
-
-            deviceId_ = input.readUInt32();
-            break;
-          }
-          case 16: {
-
-            resourceId_ = input.readUInt32();
-            break;
-          }
-          case 26: {
-            java.lang.String s = input.readStringRequireUtf8();
-
-            name_ = s;
-            break;
-          }
-          case 72: {
-
-            timestampPs_ = input.readUInt64();
-            break;
-          }
-          case 80: {
-
-            durationPs_ = input.readUInt64();
-            break;
-          }
-          case 90: {
-            if (!((mutable_bitField0_ & 0x00000001) != 0)) {
-              args_ = com.google.protobuf.MapField.newMapField(
-                  ArgsDefaultEntryHolder.defaultEntry);
-              mutable_bitField0_ |= 0x00000001;
-            }
-            com.google.protobuf.MapEntry<java.lang.String, java.lang.String>
-            args__ = input.readMessage(
-                ArgsDefaultEntryHolder.defaultEntry.getParserForType(), extensionRegistry);
-            args_.getMutableMap().put(
-                args__.getKey(), args__.getValue());
-            break;
-          }
-          default: {
-            if (!parseUnknownField(
-                input, unknownFields, extensionRegistry, tag)) {
-              done = true;
-            }
-            break;
-          }
-        }
-      }
-    } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-      throw e.setUnfinishedMessage(this);
-    } catch (java.io.IOException e) {
-      throw new com.google.protobuf.InvalidProtocolBufferException(
-          e).setUnfinishedMessage(this);
-    } finally {
-      this.unknownFields = unknownFields.build();
-      makeExtensionsImmutable();
-    }
-  }
-  public static final com.google.protobuf.Descriptors.Descriptor
-      getDescriptor() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_TraceEvent_descriptor;
-  }
-
-  @SuppressWarnings({"rawtypes"})
-  @java.lang.Override
-  protected com.google.protobuf.MapField internalGetMapField(
-      int number) {
-    switch (number) {
-      case 11:
-        return internalGetArgs();
-      default:
-        throw new RuntimeException(
-            "Invalid map field number: " + number);
-    }
-  }
-  @java.lang.Override
-  protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internalGetFieldAccessorTable() {
-    return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_TraceEvent_fieldAccessorTable
-        .ensureFieldAccessorsInitialized(
-            org.tensorflow.proto.framework.TraceEvent.class, org.tensorflow.proto.framework.TraceEvent.Builder.class);
-  }
-
-  public static final int DEVICE_ID_FIELD_NUMBER = 1;
-  private int deviceId_;
-  /**
-   * <pre>
-   * The id of the device that this event occurred on. The full dataset should
-   * have this device present in the Trace object.
-   * </pre>
-   *
-   * <code>uint32 device_id = 1;</code>
-   */
-  public int getDeviceId() {
-    return deviceId_;
-  }
-
-  public static final int RESOURCE_ID_FIELD_NUMBER = 2;
-  private int resourceId_;
-  /**
-   * <pre>
-   * The id of the resource that this event occurred on. The full dataset should
-   * have this resource present in the Device object of the Trace object. A
-   * resource_id is unique on a specific device, but not necessarily within the
-   * trace.
-   * </pre>
-   *
-   * <code>uint32 resource_id = 2;</code>
-   */
-  public int getResourceId() {
-    return resourceId_;
-  }
-
-  public static final int NAME_FIELD_NUMBER = 3;
-  private volatile java.lang.Object name_;
-  /**
-   * <pre>
-   * The name of this trace event.
-   * </pre>
-   *
-   * <code>string name = 3;</code>
-   */
-  public java.lang.String getName() {
-    java.lang.Object ref = name_;
-    if (ref instanceof java.lang.String) {
-      return (java.lang.String) ref;
-    } else {
-      com.google.protobuf.ByteString bs = 
-          (com.google.protobuf.ByteString) ref;
-      java.lang.String s = bs.toStringUtf8();
-      name_ = s;
-      return s;
-    }
-  }
-  /**
-   * <pre>
-   * The name of this trace event.
-   * </pre>
-   *
-   * <code>string name = 3;</code>
-   */
-  public com.google.protobuf.ByteString
-      getNameBytes() {
-    java.lang.Object ref = name_;
-    if (ref instanceof java.lang.String) {
-      com.google.protobuf.ByteString b = 
-          com.google.protobuf.ByteString.copyFromUtf8(
-              (java.lang.String) ref);
-      name_ = b;
-      return b;
-    } else {
-      return (com.google.protobuf.ByteString) ref;
-    }
-  }
-
-  public static final int TIMESTAMP_PS_FIELD_NUMBER = 9;
-  private long timestampPs_;
-  /**
-   * <pre>
-   * The timestamp that this event occurred at (in picos since tracing started).
-   * </pre>
-   *
-   * <code>uint64 timestamp_ps = 9;</code>
-   */
-  public long getTimestampPs() {
-    return timestampPs_;
-  }
-
-  public static final int DURATION_PS_FIELD_NUMBER = 10;
-  private long durationPs_;
-  /**
-   * <pre>
-   * The duration of the event in picoseconds if applicable.
-   * Events without duration are called instant events.
-   * </pre>
-   *
-   * <code>uint64 duration_ps = 10;</code>
-   */
-  public long getDurationPs() {
-    return durationPs_;
-  }
-
-  public static final int ARGS_FIELD_NUMBER = 11;
-  private static final class ArgsDefaultEntryHolder {
-    static final com.google.protobuf.MapEntry<
-        java.lang.String, java.lang.String> defaultEntry =
-            com.google.protobuf.MapEntry
-            .<java.lang.String, java.lang.String>newDefaultInstance(
-                org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_TraceEvent_ArgsEntry_descriptor, 
-                com.google.protobuf.WireFormat.FieldType.STRING,
-                "",
-                com.google.protobuf.WireFormat.FieldType.STRING,
-                "");
-  }
-  private com.google.protobuf.MapField<
-      java.lang.String, java.lang.String> args_;
-  private com.google.protobuf.MapField<java.lang.String, java.lang.String>
-  internalGetArgs() {
-    if (args_ == null) {
-      return com.google.protobuf.MapField.emptyMapField(
-          ArgsDefaultEntryHolder.defaultEntry);
-    }
-    return args_;
-  }
-
-  public int getArgsCount() {
-    return internalGetArgs().getMap().size();
-  }
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-
-  public boolean containsArgs(
-      java.lang.String key) {
-    if (key == null) { throw new java.lang.NullPointerException(); }
-    return internalGetArgs().getMap().containsKey(key);
-  }
-  /**
-   * Use {@link #getArgsMap()} instead.
-   */
-  @java.lang.Deprecated
-  public java.util.Map<java.lang.String, java.lang.String> getArgs() {
-    return getArgsMap();
-  }
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-
-  public java.util.Map<java.lang.String, java.lang.String> getArgsMap() {
-    return internalGetArgs().getMap();
-  }
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-
-  public java.lang.String getArgsOrDefault(
-      java.lang.String key,
-      java.lang.String defaultValue) {
-    if (key == null) { throw new java.lang.NullPointerException(); }
-    java.util.Map<java.lang.String, java.lang.String> map =
-        internalGetArgs().getMap();
-    return map.containsKey(key) ? map.get(key) : defaultValue;
-  }
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-
-  public java.lang.String getArgsOrThrow(
-      java.lang.String key) {
-    if (key == null) { throw new java.lang.NullPointerException(); }
-    java.util.Map<java.lang.String, java.lang.String> map =
-        internalGetArgs().getMap();
-    if (!map.containsKey(key)) {
-      throw new java.lang.IllegalArgumentException();
-    }
-    return map.get(key);
-  }
-
-  private byte memoizedIsInitialized = -1;
-  @java.lang.Override
-  public final boolean isInitialized() {
-    byte isInitialized = memoizedIsInitialized;
-    if (isInitialized == 1) return true;
-    if (isInitialized == 0) return false;
-
-    memoizedIsInitialized = 1;
-    return true;
-  }
-
-  @java.lang.Override
-  public void writeTo(com.google.protobuf.CodedOutputStream output)
-                      throws java.io.IOException {
-    if (deviceId_ != 0) {
-      output.writeUInt32(1, deviceId_);
-    }
-    if (resourceId_ != 0) {
-      output.writeUInt32(2, resourceId_);
-    }
-    if (!getNameBytes().isEmpty()) {
-      com.google.protobuf.GeneratedMessageV3.writeString(output, 3, name_);
-    }
-    if (timestampPs_ != 0L) {
-      output.writeUInt64(9, timestampPs_);
-    }
-    if (durationPs_ != 0L) {
-      output.writeUInt64(10, durationPs_);
-    }
-    com.google.protobuf.GeneratedMessageV3
-      .serializeStringMapTo(
-        output,
-        internalGetArgs(),
-        ArgsDefaultEntryHolder.defaultEntry,
-        11);
-    unknownFields.writeTo(output);
-  }
-
-  @java.lang.Override
-  public int getSerializedSize() {
-    int size = memoizedSize;
-    if (size != -1) return size;
-
-    size = 0;
-    if (deviceId_ != 0) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeUInt32Size(1, deviceId_);
-    }
-    if (resourceId_ != 0) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeUInt32Size(2, resourceId_);
-    }
-    if (!getNameBytes().isEmpty()) {
-      size += com.google.protobuf.GeneratedMessageV3.computeStringSize(3, name_);
-    }
-    if (timestampPs_ != 0L) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeUInt64Size(9, timestampPs_);
-    }
-    if (durationPs_ != 0L) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeUInt64Size(10, durationPs_);
-    }
-    for (java.util.Map.Entry<java.lang.String, java.lang.String> entry
-         : internalGetArgs().getMap().entrySet()) {
-      com.google.protobuf.MapEntry<java.lang.String, java.lang.String>
-      args__ = ArgsDefaultEntryHolder.defaultEntry.newBuilderForType()
-          .setKey(entry.getKey())
-          .setValue(entry.getValue())
-          .build();
-      size += com.google.protobuf.CodedOutputStream
-          .computeMessageSize(11, args__);
-    }
-    size += unknownFields.getSerializedSize();
-    memoizedSize = size;
-    return size;
-  }
-
-  @java.lang.Override
-  public boolean equals(final java.lang.Object obj) {
-    if (obj == this) {
-     return true;
-    }
-    if (!(obj instanceof org.tensorflow.proto.framework.TraceEvent)) {
-      return super.equals(obj);
-    }
-    org.tensorflow.proto.framework.TraceEvent other = (org.tensorflow.proto.framework.TraceEvent) obj;
-
-    if (getDeviceId()
-        != other.getDeviceId()) return false;
-    if (getResourceId()
-        != other.getResourceId()) return false;
-    if (!getName()
-        .equals(other.getName())) return false;
-    if (getTimestampPs()
-        != other.getTimestampPs()) return false;
-    if (getDurationPs()
-        != other.getDurationPs()) return false;
-    if (!internalGetArgs().equals(
-        other.internalGetArgs())) return false;
-    if (!unknownFields.equals(other.unknownFields)) return false;
-    return true;
-  }
-
-  @java.lang.Override
-  public int hashCode() {
-    if (memoizedHashCode != 0) {
-      return memoizedHashCode;
-    }
-    int hash = 41;
-    hash = (19 * hash) + getDescriptor().hashCode();
-    hash = (37 * hash) + DEVICE_ID_FIELD_NUMBER;
-    hash = (53 * hash) + getDeviceId();
-    hash = (37 * hash) + RESOURCE_ID_FIELD_NUMBER;
-    hash = (53 * hash) + getResourceId();
-    hash = (37 * hash) + NAME_FIELD_NUMBER;
-    hash = (53 * hash) + getName().hashCode();
-    hash = (37 * hash) + TIMESTAMP_PS_FIELD_NUMBER;
-    hash = (53 * hash) + com.google.protobuf.Internal.hashLong(
-        getTimestampPs());
-    hash = (37 * hash) + DURATION_PS_FIELD_NUMBER;
-    hash = (53 * hash) + com.google.protobuf.Internal.hashLong(
-        getDurationPs());
-    if (!internalGetArgs().getMap().isEmpty()) {
-      hash = (37 * hash) + ARGS_FIELD_NUMBER;
-      hash = (53 * hash) + internalGetArgs().hashCode();
-    }
-    hash = (29 * hash) + unknownFields.hashCode();
-    memoizedHashCode = hash;
-    return hash;
-  }
-
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      java.nio.ByteBuffer data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      java.nio.ByteBuffer data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      com.google.protobuf.ByteString data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      com.google.protobuf.ByteString data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(byte[] data)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      byte[] data,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws com.google.protobuf.InvalidProtocolBufferException {
-    return PARSER.parseFrom(data, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseDelimitedFrom(java.io.InputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseDelimitedFrom(
-      java.io.InputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseDelimitedWithIOException(PARSER, input, extensionRegistry);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      com.google.protobuf.CodedInputStream input)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input);
-  }
-  public static org.tensorflow.proto.framework.TraceEvent parseFrom(
-      com.google.protobuf.CodedInputStream input,
-      com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-      throws java.io.IOException {
-    return com.google.protobuf.GeneratedMessageV3
-        .parseWithIOException(PARSER, input, extensionRegistry);
-  }
-
-  @java.lang.Override
-  public Builder newBuilderForType() { return newBuilder(); }
-  public static Builder newBuilder() {
-    return DEFAULT_INSTANCE.toBuilder();
-  }
-  public static Builder newBuilder(org.tensorflow.proto.framework.TraceEvent prototype) {
-    return DEFAULT_INSTANCE.toBuilder().mergeFrom(prototype);
-  }
-  @java.lang.Override
-  public Builder toBuilder() {
-    return this == DEFAULT_INSTANCE
-        ? new Builder() : new Builder().mergeFrom(this);
-  }
-
-  @java.lang.Override
-  protected Builder newBuilderForType(
-      com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-    Builder builder = new Builder(parent);
-    return builder;
-  }
-  /**
-   * Protobuf type {@code tensorflow.profiler.TraceEvent}
-   */
-  public static final class Builder extends
-      com.google.protobuf.GeneratedMessageV3.Builder<Builder> implements
-      // @@protoc_insertion_point(builder_implements:tensorflow.profiler.TraceEvent)
-      org.tensorflow.proto.framework.TraceEventOrBuilder {
-    public static final com.google.protobuf.Descriptors.Descriptor
-        getDescriptor() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_TraceEvent_descriptor;
-    }
-
-    @SuppressWarnings({"rawtypes"})
-    protected com.google.protobuf.MapField internalGetMapField(
-        int number) {
-      switch (number) {
-        case 11:
-          return internalGetArgs();
-        default:
-          throw new RuntimeException(
-              "Invalid map field number: " + number);
-      }
-    }
-    @SuppressWarnings({"rawtypes"})
-    protected com.google.protobuf.MapField internalGetMutableMapField(
-        int number) {
-      switch (number) {
-        case 11:
-          return internalGetMutableArgs();
-        default:
-          throw new RuntimeException(
-              "Invalid map field number: " + number);
-      }
-    }
-    @java.lang.Override
-    protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-        internalGetFieldAccessorTable() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_TraceEvent_fieldAccessorTable
-          .ensureFieldAccessorsInitialized(
-              org.tensorflow.proto.framework.TraceEvent.class, org.tensorflow.proto.framework.TraceEvent.Builder.class);
-    }
-
-    // Construct using org.tensorflow.proto.framework.TraceEvent.newBuilder()
-    private Builder() {
-      maybeForceBuilderInitialization();
-    }
-
-    private Builder(
-        com.google.protobuf.GeneratedMessageV3.BuilderParent parent) {
-      super(parent);
-      maybeForceBuilderInitialization();
-    }
-    private void maybeForceBuilderInitialization() {
-      if (com.google.protobuf.GeneratedMessageV3
-              .alwaysUseFieldBuilders) {
-      }
-    }
-    @java.lang.Override
-    public Builder clear() {
-      super.clear();
-      deviceId_ = 0;
-
-      resourceId_ = 0;
-
-      name_ = "";
-
-      timestampPs_ = 0L;
-
-      durationPs_ = 0L;
-
-      internalGetMutableArgs().clear();
-      return this;
-    }
-
-    @java.lang.Override
-    public com.google.protobuf.Descriptors.Descriptor
-        getDescriptorForType() {
-      return org.tensorflow.proto.framework.TraceEventsProtos.internal_static_tensorflow_profiler_TraceEvent_descriptor;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.TraceEvent getDefaultInstanceForType() {
-      return org.tensorflow.proto.framework.TraceEvent.getDefaultInstance();
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.TraceEvent build() {
-      org.tensorflow.proto.framework.TraceEvent result = buildPartial();
-      if (!result.isInitialized()) {
-        throw newUninitializedMessageException(result);
-      }
-      return result;
-    }
-
-    @java.lang.Override
-    public org.tensorflow.proto.framework.TraceEvent buildPartial() {
-      org.tensorflow.proto.framework.TraceEvent result = new org.tensorflow.proto.framework.TraceEvent(this);
-      int from_bitField0_ = bitField0_;
-      result.deviceId_ = deviceId_;
-      result.resourceId_ = resourceId_;
-      result.name_ = name_;
-      result.timestampPs_ = timestampPs_;
-      result.durationPs_ = durationPs_;
-      result.args_ = internalGetArgs();
-      result.args_.makeImmutable();
-      onBuilt();
-      return result;
-    }
-
-    @java.lang.Override
-    public Builder clone() {
-      return super.clone();
-    }
-    @java.lang.Override
-    public Builder setField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.setField(field, value);
-    }
-    @java.lang.Override
-    public Builder clearField(
-        com.google.protobuf.Descriptors.FieldDescriptor field) {
-      return super.clearField(field);
-    }
-    @java.lang.Override
-    public Builder clearOneof(
-        com.google.protobuf.Descriptors.OneofDescriptor oneof) {
-      return super.clearOneof(oneof);
-    }
-    @java.lang.Override
-    public Builder setRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        int index, java.lang.Object value) {
-      return super.setRepeatedField(field, index, value);
-    }
-    @java.lang.Override
-    public Builder addRepeatedField(
-        com.google.protobuf.Descriptors.FieldDescriptor field,
-        java.lang.Object value) {
-      return super.addRepeatedField(field, value);
-    }
-    @java.lang.Override
-    public Builder mergeFrom(com.google.protobuf.Message other) {
-      if (other instanceof org.tensorflow.proto.framework.TraceEvent) {
-        return mergeFrom((org.tensorflow.proto.framework.TraceEvent)other);
-      } else {
-        super.mergeFrom(other);
-        return this;
-      }
-    }
-
-    public Builder mergeFrom(org.tensorflow.proto.framework.TraceEvent other) {
-      if (other == org.tensorflow.proto.framework.TraceEvent.getDefaultInstance()) return this;
-      if (other.getDeviceId() != 0) {
-        setDeviceId(other.getDeviceId());
-      }
-      if (other.getResourceId() != 0) {
-        setResourceId(other.getResourceId());
-      }
-      if (!other.getName().isEmpty()) {
-        name_ = other.name_;
-        onChanged();
-      }
-      if (other.getTimestampPs() != 0L) {
-        setTimestampPs(other.getTimestampPs());
-      }
-      if (other.getDurationPs() != 0L) {
-        setDurationPs(other.getDurationPs());
-      }
-      internalGetMutableArgs().mergeFrom(
-          other.internalGetArgs());
-      this.mergeUnknownFields(other.unknownFields);
-      onChanged();
-      return this;
-    }
-
-    @java.lang.Override
-    public final boolean isInitialized() {
-      return true;
-    }
-
-    @java.lang.Override
-    public Builder mergeFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws java.io.IOException {
-      org.tensorflow.proto.framework.TraceEvent parsedMessage = null;
-      try {
-        parsedMessage = PARSER.parsePartialFrom(input, extensionRegistry);
-      } catch (com.google.protobuf.InvalidProtocolBufferException e) {
-        parsedMessage = (org.tensorflow.proto.framework.TraceEvent) e.getUnfinishedMessage();
-        throw e.unwrapIOException();
-      } finally {
-        if (parsedMessage != null) {
-          mergeFrom(parsedMessage);
-        }
-      }
-      return this;
-    }
-    private int bitField0_;
-
-    private int deviceId_ ;
-    /**
-     * <pre>
-     * The id of the device that this event occurred on. The full dataset should
-     * have this device present in the Trace object.
-     * </pre>
-     *
-     * <code>uint32 device_id = 1;</code>
-     */
-    public int getDeviceId() {
-      return deviceId_;
-    }
-    /**
-     * <pre>
-     * The id of the device that this event occurred on. The full dataset should
-     * have this device present in the Trace object.
-     * </pre>
-     *
-     * <code>uint32 device_id = 1;</code>
-     */
-    public Builder setDeviceId(int value) {
-      
-      deviceId_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The id of the device that this event occurred on. The full dataset should
-     * have this device present in the Trace object.
-     * </pre>
-     *
-     * <code>uint32 device_id = 1;</code>
-     */
-    public Builder clearDeviceId() {
-      
-      deviceId_ = 0;
-      onChanged();
-      return this;
-    }
-
-    private int resourceId_ ;
-    /**
-     * <pre>
-     * The id of the resource that this event occurred on. The full dataset should
-     * have this resource present in the Device object of the Trace object. A
-     * resource_id is unique on a specific device, but not necessarily within the
-     * trace.
-     * </pre>
-     *
-     * <code>uint32 resource_id = 2;</code>
-     */
-    public int getResourceId() {
-      return resourceId_;
-    }
-    /**
-     * <pre>
-     * The id of the resource that this event occurred on. The full dataset should
-     * have this resource present in the Device object of the Trace object. A
-     * resource_id is unique on a specific device, but not necessarily within the
-     * trace.
-     * </pre>
-     *
-     * <code>uint32 resource_id = 2;</code>
-     */
-    public Builder setResourceId(int value) {
-      
-      resourceId_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The id of the resource that this event occurred on. The full dataset should
-     * have this resource present in the Device object of the Trace object. A
-     * resource_id is unique on a specific device, but not necessarily within the
-     * trace.
-     * </pre>
-     *
-     * <code>uint32 resource_id = 2;</code>
-     */
-    public Builder clearResourceId() {
-      
-      resourceId_ = 0;
-      onChanged();
-      return this;
-    }
-
-    private java.lang.Object name_ = "";
-    /**
-     * <pre>
-     * The name of this trace event.
-     * </pre>
-     *
-     * <code>string name = 3;</code>
-     */
-    public java.lang.String getName() {
-      java.lang.Object ref = name_;
-      if (!(ref instanceof java.lang.String)) {
-        com.google.protobuf.ByteString bs =
-            (com.google.protobuf.ByteString) ref;
-        java.lang.String s = bs.toStringUtf8();
-        name_ = s;
-        return s;
-      } else {
-        return (java.lang.String) ref;
-      }
-    }
-    /**
-     * <pre>
-     * The name of this trace event.
-     * </pre>
-     *
-     * <code>string name = 3;</code>
-     */
-    public com.google.protobuf.ByteString
-        getNameBytes() {
-      java.lang.Object ref = name_;
-      if (ref instanceof String) {
-        com.google.protobuf.ByteString b = 
-            com.google.protobuf.ByteString.copyFromUtf8(
-                (java.lang.String) ref);
-        name_ = b;
-        return b;
-      } else {
-        return (com.google.protobuf.ByteString) ref;
-      }
-    }
-    /**
-     * <pre>
-     * The name of this trace event.
-     * </pre>
-     *
-     * <code>string name = 3;</code>
-     */
-    public Builder setName(
-        java.lang.String value) {
-      if (value == null) {
-    throw new NullPointerException();
-  }
-  
-      name_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The name of this trace event.
-     * </pre>
-     *
-     * <code>string name = 3;</code>
-     */
-    public Builder clearName() {
-      
-      name_ = getDefaultInstance().getName();
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The name of this trace event.
-     * </pre>
-     *
-     * <code>string name = 3;</code>
-     */
-    public Builder setNameBytes(
-        com.google.protobuf.ByteString value) {
-      if (value == null) {
-    throw new NullPointerException();
-  }
-  checkByteStringIsUtf8(value);
-      
-      name_ = value;
-      onChanged();
-      return this;
-    }
-
-    private long timestampPs_ ;
-    /**
-     * <pre>
-     * The timestamp that this event occurred at (in picos since tracing started).
-     * </pre>
-     *
-     * <code>uint64 timestamp_ps = 9;</code>
-     */
-    public long getTimestampPs() {
-      return timestampPs_;
-    }
-    /**
-     * <pre>
-     * The timestamp that this event occurred at (in picos since tracing started).
-     * </pre>
-     *
-     * <code>uint64 timestamp_ps = 9;</code>
-     */
-    public Builder setTimestampPs(long value) {
-      
-      timestampPs_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The timestamp that this event occurred at (in picos since tracing started).
-     * </pre>
-     *
-     * <code>uint64 timestamp_ps = 9;</code>
-     */
-    public Builder clearTimestampPs() {
-      
-      timestampPs_ = 0L;
-      onChanged();
-      return this;
-    }
-
-    private long durationPs_ ;
-    /**
-     * <pre>
-     * The duration of the event in picoseconds if applicable.
-     * Events without duration are called instant events.
-     * </pre>
-     *
-     * <code>uint64 duration_ps = 10;</code>
-     */
-    public long getDurationPs() {
-      return durationPs_;
-    }
-    /**
-     * <pre>
-     * The duration of the event in picoseconds if applicable.
-     * Events without duration are called instant events.
-     * </pre>
-     *
-     * <code>uint64 duration_ps = 10;</code>
-     */
-    public Builder setDurationPs(long value) {
-      
-      durationPs_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * The duration of the event in picoseconds if applicable.
-     * Events without duration are called instant events.
-     * </pre>
-     *
-     * <code>uint64 duration_ps = 10;</code>
-     */
-    public Builder clearDurationPs() {
-      
-      durationPs_ = 0L;
-      onChanged();
-      return this;
-    }
-
-    private com.google.protobuf.MapField<
-        java.lang.String, java.lang.String> args_;
-    private com.google.protobuf.MapField<java.lang.String, java.lang.String>
-    internalGetArgs() {
-      if (args_ == null) {
-        return com.google.protobuf.MapField.emptyMapField(
-            ArgsDefaultEntryHolder.defaultEntry);
-      }
-      return args_;
-    }
-    private com.google.protobuf.MapField<java.lang.String, java.lang.String>
-    internalGetMutableArgs() {
-      onChanged();;
-      if (args_ == null) {
-        args_ = com.google.protobuf.MapField.newMapField(
-            ArgsDefaultEntryHolder.defaultEntry);
-      }
-      if (!args_.isMutable()) {
-        args_ = args_.copy();
-      }
-      return args_;
-    }
-
-    public int getArgsCount() {
-      return internalGetArgs().getMap().size();
-    }
-    /**
-     * <pre>
-     * Extra arguments that will be displayed in trace view.
-     * </pre>
-     *
-     * <code>map&lt;string, string&gt; args = 11;</code>
-     */
-
-    public boolean containsArgs(
-        java.lang.String key) {
-      if (key == null) { throw new java.lang.NullPointerException(); }
-      return internalGetArgs().getMap().containsKey(key);
-    }
-    /**
-     * Use {@link #getArgsMap()} instead.
-     */
-    @java.lang.Deprecated
-    public java.util.Map<java.lang.String, java.lang.String> getArgs() {
-      return getArgsMap();
-    }
-    /**
-     * <pre>
-     * Extra arguments that will be displayed in trace view.
-     * </pre>
-     *
-     * <code>map&lt;string, string&gt; args = 11;</code>
-     */
-
-    public java.util.Map<java.lang.String, java.lang.String> getArgsMap() {
-      return internalGetArgs().getMap();
-    }
-    /**
-     * <pre>
-     * Extra arguments that will be displayed in trace view.
-     * </pre>
-     *
-     * <code>map&lt;string, string&gt; args = 11;</code>
-     */
-
-    public java.lang.String getArgsOrDefault(
-        java.lang.String key,
-        java.lang.String defaultValue) {
-      if (key == null) { throw new java.lang.NullPointerException(); }
-      java.util.Map<java.lang.String, java.lang.String> map =
-          internalGetArgs().getMap();
-      return map.containsKey(key) ? map.get(key) : defaultValue;
-    }
-    /**
-     * <pre>
-     * Extra arguments that will be displayed in trace view.
-     * </pre>
-     *
-     * <code>map&lt;string, string&gt; args = 11;</code>
-     */
-
-    public java.lang.String getArgsOrThrow(
-        java.lang.String key) {
-      if (key == null) { throw new java.lang.NullPointerException(); }
-      java.util.Map<java.lang.String, java.lang.String> map =
-          internalGetArgs().getMap();
-      if (!map.containsKey(key)) {
-        throw new java.lang.IllegalArgumentException();
-      }
-      return map.get(key);
-    }
-
-    public Builder clearArgs() {
-      internalGetMutableArgs().getMutableMap()
-          .clear();
-      return this;
-    }
-    /**
-     * <pre>
-     * Extra arguments that will be displayed in trace view.
-     * </pre>
-     *
-     * <code>map&lt;string, string&gt; args = 11;</code>
-     */
-
-    public Builder removeArgs(
-        java.lang.String key) {
-      if (key == null) { throw new java.lang.NullPointerException(); }
-      internalGetMutableArgs().getMutableMap()
-          .remove(key);
-      return this;
-    }
-    /**
-     * Use alternate mutation accessors instead.
-     */
-    @java.lang.Deprecated
-    public java.util.Map<java.lang.String, java.lang.String>
-    getMutableArgs() {
-      return internalGetMutableArgs().getMutableMap();
-    }
-    /**
-     * <pre>
-     * Extra arguments that will be displayed in trace view.
-     * </pre>
-     *
-     * <code>map&lt;string, string&gt; args = 11;</code>
-     */
-    public Builder putArgs(
-        java.lang.String key,
-        java.lang.String value) {
-      if (key == null) { throw new java.lang.NullPointerException(); }
-      if (value == null) { throw new java.lang.NullPointerException(); }
-      internalGetMutableArgs().getMutableMap()
-          .put(key, value);
-      return this;
-    }
-    /**
-     * <pre>
-     * Extra arguments that will be displayed in trace view.
-     * </pre>
-     *
-     * <code>map&lt;string, string&gt; args = 11;</code>
-     */
-
-    public Builder putAllArgs(
-        java.util.Map<java.lang.String, java.lang.String> values) {
-      internalGetMutableArgs().getMutableMap()
-          .putAll(values);
-      return this;
-    }
-    @java.lang.Override
-    public final Builder setUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.setUnknownFields(unknownFields);
-    }
-
-    @java.lang.Override
-    public final Builder mergeUnknownFields(
-        final com.google.protobuf.UnknownFieldSet unknownFields) {
-      return super.mergeUnknownFields(unknownFields);
-    }
-
-
-    // @@protoc_insertion_point(builder_scope:tensorflow.profiler.TraceEvent)
-  }
-
-  // @@protoc_insertion_point(class_scope:tensorflow.profiler.TraceEvent)
-  private static final org.tensorflow.proto.framework.TraceEvent DEFAULT_INSTANCE;
-  static {
-    DEFAULT_INSTANCE = new org.tensorflow.proto.framework.TraceEvent();
-  }
-
-  public static org.tensorflow.proto.framework.TraceEvent getDefaultInstance() {
-    return DEFAULT_INSTANCE;
-  }
-
-  private static final com.google.protobuf.Parser<TraceEvent>
-      PARSER = new com.google.protobuf.AbstractParser<TraceEvent>() {
-    @java.lang.Override
-    public TraceEvent parsePartialFrom(
-        com.google.protobuf.CodedInputStream input,
-        com.google.protobuf.ExtensionRegistryLite extensionRegistry)
-        throws com.google.protobuf.InvalidProtocolBufferException {
-      return new TraceEvent(input, extensionRegistry);
-    }
-  };
-
-  public static com.google.protobuf.Parser<TraceEvent> parser() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public com.google.protobuf.Parser<TraceEvent> getParserForType() {
-    return PARSER;
-  }
-
-  @java.lang.Override
-  public org.tensorflow.proto.framework.TraceEvent getDefaultInstanceForType() {
-    return DEFAULT_INSTANCE;
-  }
-
-}
-
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEventOrBuilder.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEventOrBuilder.java
deleted file mode 100644
index 2bebd63d083..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEventOrBuilder.java
+++ /dev/null
@@ -1,122 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-public interface TraceEventOrBuilder extends
-    // @@protoc_insertion_point(interface_extends:tensorflow.profiler.TraceEvent)
-    com.google.protobuf.MessageOrBuilder {
-
-  /**
-   * <pre>
-   * The id of the device that this event occurred on. The full dataset should
-   * have this device present in the Trace object.
-   * </pre>
-   *
-   * <code>uint32 device_id = 1;</code>
-   */
-  int getDeviceId();
-
-  /**
-   * <pre>
-   * The id of the resource that this event occurred on. The full dataset should
-   * have this resource present in the Device object of the Trace object. A
-   * resource_id is unique on a specific device, but not necessarily within the
-   * trace.
-   * </pre>
-   *
-   * <code>uint32 resource_id = 2;</code>
-   */
-  int getResourceId();
-
-  /**
-   * <pre>
-   * The name of this trace event.
-   * </pre>
-   *
-   * <code>string name = 3;</code>
-   */
-  java.lang.String getName();
-  /**
-   * <pre>
-   * The name of this trace event.
-   * </pre>
-   *
-   * <code>string name = 3;</code>
-   */
-  com.google.protobuf.ByteString
-      getNameBytes();
-
-  /**
-   * <pre>
-   * The timestamp that this event occurred at (in picos since tracing started).
-   * </pre>
-   *
-   * <code>uint64 timestamp_ps = 9;</code>
-   */
-  long getTimestampPs();
-
-  /**
-   * <pre>
-   * The duration of the event in picoseconds if applicable.
-   * Events without duration are called instant events.
-   * </pre>
-   *
-   * <code>uint64 duration_ps = 10;</code>
-   */
-  long getDurationPs();
-
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-  int getArgsCount();
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-  boolean containsArgs(
-      java.lang.String key);
-  /**
-   * Use {@link #getArgsMap()} instead.
-   */
-  @java.lang.Deprecated
-  java.util.Map<java.lang.String, java.lang.String>
-  getArgs();
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-  java.util.Map<java.lang.String, java.lang.String>
-  getArgsMap();
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-
-  java.lang.String getArgsOrDefault(
-      java.lang.String key,
-      java.lang.String defaultValue);
-  /**
-   * <pre>
-   * Extra arguments that will be displayed in trace view.
-   * </pre>
-   *
-   * <code>map&lt;string, string&gt; args = 11;</code>
-   */
-
-  java.lang.String getArgsOrThrow(
-      java.lang.String key);
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEventsProtos.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEventsProtos.java
deleted file mode 100644
index 4e2e8c8d80a..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceEventsProtos.java
+++ /dev/null
@@ -1,133 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-public final class TraceEventsProtos {
-  private TraceEventsProtos() {}
-  public static void registerAllExtensions(
-      com.google.protobuf.ExtensionRegistryLite registry) {
-  }
-
-  public static void registerAllExtensions(
-      com.google.protobuf.ExtensionRegistry registry) {
-    registerAllExtensions(
-        (com.google.protobuf.ExtensionRegistryLite) registry);
-  }
-  static final com.google.protobuf.Descriptors.Descriptor
-    internal_static_tensorflow_profiler_Trace_descriptor;
-  static final 
-    com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internal_static_tensorflow_profiler_Trace_fieldAccessorTable;
-  static final com.google.protobuf.Descriptors.Descriptor
-    internal_static_tensorflow_profiler_Trace_DevicesEntry_descriptor;
-  static final 
-    com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internal_static_tensorflow_profiler_Trace_DevicesEntry_fieldAccessorTable;
-  static final com.google.protobuf.Descriptors.Descriptor
-    internal_static_tensorflow_profiler_Device_descriptor;
-  static final 
-    com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internal_static_tensorflow_profiler_Device_fieldAccessorTable;
-  static final com.google.protobuf.Descriptors.Descriptor
-    internal_static_tensorflow_profiler_Device_ResourcesEntry_descriptor;
-  static final 
-    com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internal_static_tensorflow_profiler_Device_ResourcesEntry_fieldAccessorTable;
-  static final com.google.protobuf.Descriptors.Descriptor
-    internal_static_tensorflow_profiler_Resource_descriptor;
-  static final 
-    com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internal_static_tensorflow_profiler_Resource_fieldAccessorTable;
-  static final com.google.protobuf.Descriptors.Descriptor
-    internal_static_tensorflow_profiler_TraceEvent_descriptor;
-  static final 
-    com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internal_static_tensorflow_profiler_TraceEvent_fieldAccessorTable;
-  static final com.google.protobuf.Descriptors.Descriptor
-    internal_static_tensorflow_profiler_TraceEvent_ArgsEntry_descriptor;
-  static final 
-    com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
-      internal_static_tensorflow_profiler_TraceEvent_ArgsEntry_fieldAccessorTable;
-
-  public static com.google.protobuf.Descriptors.FileDescriptor
-      getDescriptor() {
-    return descriptor;
-  }
-  private static  com.google.protobuf.Descriptors.FileDescriptor
-      descriptor;
-  static {
-    java.lang.String[] descriptorData = {
-      "\n+tensorflow/core/protobuf/trace_events." +
-      "proto\022\023tensorflow.profiler\"\305\001\n\005Trace\0228\n\007" +
-      "devices\030\001 \003(\0132\'.tensorflow.profiler.Trac" +
-      "e.DevicesEntry\0225\n\014trace_events\030\004 \003(\0132\037.t" +
-      "ensorflow.profiler.TraceEvent\032K\n\014Devices" +
-      "Entry\022\013\n\003key\030\001 \001(\r\022*\n\005value\030\002 \001(\0132\033.tens" +
-      "orflow.profiler.Device:\0028\001\"\271\001\n\006Device\022\014\n" +
-      "\004name\030\001 \001(\t\022\021\n\tdevice_id\030\002 \001(\r\022=\n\tresour" +
-      "ces\030\003 \003(\0132*.tensorflow.profiler.Device.R" +
-      "esourcesEntry\032O\n\016ResourcesEntry\022\013\n\003key\030\001" +
-      " \001(\r\022,\n\005value\030\002 \001(\0132\035.tensorflow.profile" +
-      "r.Resource:\0028\001\"-\n\010Resource\022\014\n\004name\030\001 \001(\t" +
-      "\022\023\n\013resource_id\030\002 \001(\r\"\323\001\n\nTraceEvent\022\021\n\t" +
-      "device_id\030\001 \001(\r\022\023\n\013resource_id\030\002 \001(\r\022\014\n\004" +
-      "name\030\003 \001(\t\022\024\n\014timestamp_ps\030\t \001(\004\022\023\n\013dura" +
-      "tion_ps\030\n \001(\004\0227\n\004args\030\013 \003(\0132).tensorflow" +
-      ".profiler.TraceEvent.ArgsEntry\032+\n\tArgsEn" +
-      "try\022\013\n\003key\030\001 \001(\t\022\r\n\005value\030\002 \001(\t:\0028\001B\202\001\n\036" +
-      "org.tensorflow.proto.frameworkB\021TraceEve" +
-      "ntsProtosP\001ZHgithub.com/tensorflow/tenso" +
-      "rflow/tensorflow/go/core/core_protos_go_" +
-      "proto\370\001\001b\006proto3"
-    };
-    descriptor = com.google.protobuf.Descriptors.FileDescriptor
-      .internalBuildGeneratedFileFrom(descriptorData,
-        new com.google.protobuf.Descriptors.FileDescriptor[] {
-        });
-    internal_static_tensorflow_profiler_Trace_descriptor =
-      getDescriptor().getMessageTypes().get(0);
-    internal_static_tensorflow_profiler_Trace_fieldAccessorTable = new
-      com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
-        internal_static_tensorflow_profiler_Trace_descriptor,
-        new java.lang.String[] { "Devices", "TraceEvents", });
-    internal_static_tensorflow_profiler_Trace_DevicesEntry_descriptor =
-      internal_static_tensorflow_profiler_Trace_descriptor.getNestedTypes().get(0);
-    internal_static_tensorflow_profiler_Trace_DevicesEntry_fieldAccessorTable = new
-      com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
-        internal_static_tensorflow_profiler_Trace_DevicesEntry_descriptor,
-        new java.lang.String[] { "Key", "Value", });
-    internal_static_tensorflow_profiler_Device_descriptor =
-      getDescriptor().getMessageTypes().get(1);
-    internal_static_tensorflow_profiler_Device_fieldAccessorTable = new
-      com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
-        internal_static_tensorflow_profiler_Device_descriptor,
-        new java.lang.String[] { "Name", "DeviceId", "Resources", });
-    internal_static_tensorflow_profiler_Device_ResourcesEntry_descriptor =
-      internal_static_tensorflow_profiler_Device_descriptor.getNestedTypes().get(0);
-    internal_static_tensorflow_profiler_Device_ResourcesEntry_fieldAccessorTable = new
-      com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
-        internal_static_tensorflow_profiler_Device_ResourcesEntry_descriptor,
-        new java.lang.String[] { "Key", "Value", });
-    internal_static_tensorflow_profiler_Resource_descriptor =
-      getDescriptor().getMessageTypes().get(2);
-    internal_static_tensorflow_profiler_Resource_fieldAccessorTable = new
-      com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
-        internal_static_tensorflow_profiler_Resource_descriptor,
-        new java.lang.String[] { "Name", "ResourceId", });
-    internal_static_tensorflow_profiler_TraceEvent_descriptor =
-      getDescriptor().getMessageTypes().get(3);
-    internal_static_tensorflow_profiler_TraceEvent_fieldAccessorTable = new
-      com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
-        internal_static_tensorflow_profiler_TraceEvent_descriptor,
-        new java.lang.String[] { "DeviceId", "ResourceId", "Name", "TimestampPs", "DurationPs", "Args", });
-    internal_static_tensorflow_profiler_TraceEvent_ArgsEntry_descriptor =
-      internal_static_tensorflow_profiler_TraceEvent_descriptor.getNestedTypes().get(0);
-    internal_static_tensorflow_profiler_TraceEvent_ArgsEntry_fieldAccessorTable = new
-      com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
-        internal_static_tensorflow_profiler_TraceEvent_ArgsEntry_descriptor,
-        new java.lang.String[] { "Key", "Value", });
-  }
-
-  // @@protoc_insertion_point(outer_class_scope)
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceOrBuilder.java b/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceOrBuilder.java
deleted file mode 100644
index 78fa25b0aa9..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/proto/framework/TraceOrBuilder.java
+++ /dev/null
@@ -1,112 +0,0 @@
-// Generated by the protocol buffer compiler.  DO NOT EDIT!
-// source: tensorflow/core/protobuf/trace_events.proto
-
-package org.tensorflow.proto.framework;
-
-public interface TraceOrBuilder extends
-    // @@protoc_insertion_point(interface_extends:tensorflow.profiler.Trace)
-    com.google.protobuf.MessageOrBuilder {
-
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-  int getDevicesCount();
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-  boolean containsDevices(
-      int key);
-  /**
-   * Use {@link #getDevicesMap()} instead.
-   */
-  @java.lang.Deprecated
-  java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device>
-  getDevices();
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-  java.util.Map<java.lang.Integer, org.tensorflow.proto.framework.Device>
-  getDevicesMap();
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-
-  org.tensorflow.proto.framework.Device getDevicesOrDefault(
-      int key,
-      org.tensorflow.proto.framework.Device defaultValue);
-  /**
-   * <pre>
-   * The devices that this trace has information about. Maps from device_id to
-   * more data about the specific device.
-   * </pre>
-   *
-   * <code>map&lt;uint32, .tensorflow.profiler.Device&gt; devices = 1;</code>
-   */
-
-  org.tensorflow.proto.framework.Device getDevicesOrThrow(
-      int key);
-
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  java.util.List<org.tensorflow.proto.framework.TraceEvent> 
-      getTraceEventsList();
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  org.tensorflow.proto.framework.TraceEvent getTraceEvents(int index);
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  int getTraceEventsCount();
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  java.util.List<? extends org.tensorflow.proto.framework.TraceEventOrBuilder> 
-      getTraceEventsOrBuilderList();
-  /**
-   * <pre>
-   * All trace events capturing in the profiling period.
-   * </pre>
-   *
-   * <code>repeated .tensorflow.profiler.TraceEvent trace_events = 4;</code>
-   */
-  org.tensorflow.proto.framework.TraceEventOrBuilder getTraceEventsOrBuilder(
-      int index);
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/AbstractOperation.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/AbstractOperation.java
index 96da6bc5ff4..80aa928a09a 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/AbstractOperation.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/AbstractOperation.java
@@ -17,6 +17,7 @@
 
 import org.bytedeco.javacpp.Pointer;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.proto.framework.DataType;
 import org.tensorflow.types.family.TType;
 
 /**
@@ -74,9 +75,9 @@ public String toString() {
    * Returns the datatype of the tensor of the {@code outputIdx}th output of this operation.
    *
    * @param outputIdx index of the output of this operation
-   * @return output tensor datatype
+   * @return datatype native code
    */
-  abstract DataType<?> dtype(int outputIdx);
+  abstract DataType dtype(int outputIdx);
 
   /**
    * Returns the tensor of the {@code outputIdx}th output of this operation.
@@ -86,5 +87,5 @@ public String toString() {
    * @param outputIdx index of the output of this operation
    * @return output tensor
    */
-  abstract Tensor<?> tensor(int outputIdx);
+  abstract TensorHandle tensor(int outputIdx);
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/AbstractTypeFactory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/AbstractTypeFactory.java
new file mode 100644
index 00000000000..a1846f3dede
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/AbstractTypeFactory.java
@@ -0,0 +1,12 @@
+package org.tensorflow;
+
+import org.tensorflow.internal.c_api.TF_Tensor;
+import org.tensorflow.types.TypeFactory;
+import org.tensorflow.types.family.TType;
+
+public abstract class AbstractTypeFactory<T extends TType> implements TypeFactory<T> {
+
+  protected static TF_Tensor getNative(TensorHandle tensorHandle) {
+    return tensorHandle.get();
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/ConcreteFunction.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/ConcreteFunction.java
index 872b4b4d16d..de516eba81f 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/ConcreteFunction.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/ConcreteFunction.java
@@ -16,14 +16,13 @@
 package org.tensorflow;
 
 import java.io.IOException;
-import java.util.List;
-import java.util.ListIterator;
-import java.util.HashMap;
 import java.util.Map;
 import java.util.function.Function;
 import org.tensorflow.op.Ops;
 import org.tensorflow.proto.framework.SignatureDef;
 import org.tensorflow.proto.framework.TensorInfo;
+import org.tensorflow.types.family.TType;
+import org.tensorflow.util.TensorMap;
 
 /**
  * A graph that can be invoked as a single function, with an input and output signature.
@@ -163,14 +162,14 @@ public Signature signature() {
    * @return output tensors resulting from the execution of the function,
    *         mapped by their signature name
    */
-  public Map<String, Tensor<?>> call(Map<String, Tensor<?>> arguments)
+  public TensorMap call(Map<String, TType> arguments)
       throws IllegalArgumentException {
 
     final SignatureDef signatureDef = signature.asSignatureDef();
     final Session.Runner runner = session.runner();
 
     signatureDef.getInputsMap().forEach((argName, t) -> {
-      Tensor<?> tensor = arguments.get(argName);
+      TType tensor = arguments.get(argName);
       if (tensor == null) {
         throw new IllegalArgumentException(String.format("Missing argument [%s]", argName));
       }
@@ -180,24 +179,7 @@ public Map<String, Tensor<?>> call(Map<String, Tensor<?>> arguments)
     Map<String, TensorInfo> outputToNode = signatureDef.getOutputsMap();
     outputToNode.values().forEach(t -> runner.fetch(t.getName()));
 
-    List<Tensor<?>> resultTensors = runner.run();
-    try {
-      ListIterator<Tensor<?>> resultTensorIter = resultTensors.listIterator();
-      Map<String, Tensor<?>> returnMap = new HashMap<String, Tensor<?>>();
-
-      // Use the output names as present in the signature definition
-      for (String nodeName: outputToNode.keySet()) {
-        returnMap.put(nodeName, resultTensorIter.next());
-      }
-      return returnMap;
-
-    } catch (Exception e) {
-      // Release tensors before throwing exception
-      for (Tensor<?> t : resultTensors) {
-        t.close();
-      }
-      throw e;
-    }
+    return runner.run().toMap(outputToNode.keySet());
   }
 
   /**
@@ -210,7 +192,7 @@ public Map<String, Tensor<?>> call(Map<String, Tensor<?>> arguments)
    * @throws IllegalArgumentException if there are multiple input or output parameters defined
    *                                  in the function
    */
-  public Tensor<?> call(Tensor<?> tensor) throws IllegalArgumentException {
+  public <T extends TType> T call(TType tensor) throws IllegalArgumentException {
     final SignatureDef signatureDef = signature.asSignatureDef();
 
     if (signatureDef.getInputsCount() != 1) {
@@ -225,7 +207,7 @@ public Tensor<?> call(Tensor<?> tensor) throws IllegalArgumentException {
     }
     String outputNodeName = signatureDef.getOutputsMap().values().iterator().next().getName();
 
-    return session.runner().feed(inputNodeName, tensor).fetch(outputNodeName).run().get(0);
+    return (T)session.runner().feed(inputNodeName, tensor).fetch(outputNodeName).run().get(0);
   }
 
   /**
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/DataType.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/DataType.java
deleted file mode 100644
index 7b76b6dd02e..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/DataType.java
+++ /dev/null
@@ -1,181 +0,0 @@
-/* Copyright 2016 The TensorFlow Authors. All Rights Reserved.
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-==============================================================================*/
-
-package org.tensorflow;
-
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.types.TBfloat16;
-import org.tensorflow.types.TBool;
-import org.tensorflow.types.TFloat16;
-import org.tensorflow.types.TFloat32;
-import org.tensorflow.types.TFloat64;
-import org.tensorflow.types.TInt32;
-import org.tensorflow.types.TInt64;
-import org.tensorflow.types.TUint8;
-import org.tensorflow.types.TString;
-import org.tensorflow.types.family.TType;
-
-/** Represents a type of elements in a {@link Tensor} */
-public final class DataType<T extends TType> {
-
-  @FunctionalInterface
-  public interface TensorMapper<T> {
-
-    /**
-     * Maps tensor memory to a data structure for manipulating elements of this type.
-     *
-     * @param nativeTensor pointer to the native tensor
-     * @param shape the shape of the tensor
-     * @return data structure of elements of this type
-     */
-    T apply(TF_Tensor nativeTensor, Shape shape);
-  }
-
-  /**
-   * Creates a new datatype
-   *
-   * @param name readable-name for this type
-   * @param value must match the corresponding TF_* value in the TensorFlow C API.
-   * @param byteSize size of an element of this type, in bytes, -1 if unknown
-   * @param <T> a tensor type
-   * @param tensorMapper method for mapping tensor memory to elements of this type
-   */
-  public static <T extends TType> DataType<T> create(
-      String name, int value, int byteSize, TensorMapper<T> tensorMapper) {
-    return new DataType<>(name, value, byteSize, tensorMapper);
-  }
-
-  /**
-   * Gets the DataType associated with the readable-name for the type
-   * <p>The name must match exactly the name used to create the desired DataType</p>
-   *
-   * @param name readable-name for the type
-   * @return the DataType
-   * @throws java.lang.IllegalArgumentException if the name is not a valid data type name
-   * @throws java.lang.NullPointerException if name is null
-   */
-  public static DataType<? extends TType> of(String name) {
-    switch (name) {
-      case TBfloat16.NAME:
-        return TBfloat16.DTYPE;
-      case TFloat16.NAME:
-        return TFloat16.DTYPE;
-      case TFloat32.NAME:
-        return TFloat32.DTYPE;
-      case TFloat64.NAME:
-        return TFloat64.DTYPE;
-      case TUint8.NAME:
-        return TUint8.DTYPE;
-      case TInt32.NAME:
-        return TInt32.DTYPE;
-      case TInt64.NAME:
-        return TInt64.DTYPE;
-      case TBool.NAME:
-        return TBool.DTYPE;
-      case TString.NAME:
-        return TString.DTYPE;
-      default:
-        throw new IllegalArgumentException(String.format("%s is an unknown DataType", name));
-    }
-  }
-
-  /** Returns true if this data type represents a floating point type */
-  public boolean isFloating() {
-    switch (this.name()) {
-      case TBfloat16.NAME:
-      case TFloat16.NAME:
-      case TFloat32.NAME:
-      case TFloat64.NAME:
-        return true;
-      default:
-        return false;
-    }
-  }
-
-  /** Returns true if this data type represents an integer type */
-  public boolean isInteger() {
-    switch (this.name()) {
-      case TInt32.NAME:
-      case TInt64.NAME:
-      case TUint8.NAME:
-        return true;
-      default:
-        return false;
-    }
-  }
-
-  /** Returns true if this data type represents a numeric type */
-  public boolean isNumeric() {
-    return isFloating() || isInteger();
-  }
-
-  /** Returns true if this data type represents a boolean type */
-  public boolean isBoolean() {
-    return this.name().equals(TBool.NAME);
-  }
-
-  /** Returns true if this data type represents a string type */
-  public boolean isString() {
-    return this.name().equals(TString.NAME);
-  }
-
-  /** Returns the size of an element of this type, in bytes, or -1 if element size is variable. */
-  public int byteSize() {
-    return byteSize;
-  }
-
-  /** Returns true if this datatype has elements of variable length */
-  public boolean isVariableLength() {
-    return byteSize == -1;
-  }
-
-  /** Returns a readable name for this type */
-  public String name() {
-    return name;
-  }
-
-  @Override
-  public String toString() {
-    return name + " (" + nativeCode + ")";
-  }
-
-  /** Returns the numeric code for this datatype, as recognized by the native library (C API) */
-  int nativeCode() {
-    return nativeCode;
-  }
-
-  /**
-   * Maps a tensor to a data structure for manipulating elements of this type.
-   *
-   * @param tensor tensor to map
-   * @return data structure of elements of this type
-   */
-  T map(Tensor<T> tensor) {
-    return tensorMapper.apply(tensor.nativeHandle(), tensor.shape());
-  }
-
-  private final int nativeCode;
-  private final int byteSize;
-  private final String name;
-  private final TensorMapper<T> tensorMapper;
-
-  private DataType(String name, int nativeCode, int byteSize, TensorMapper<T> tensorMapper) {
-    this.name = name;
-    this.nativeCode = nativeCode;
-    this.byteSize = byteSize;
-    this.tensorMapper = tensorMapper;
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/DataTypes.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/DataTypes.java
deleted file mode 100644
index 77c0de0c83f..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/DataTypes.java
+++ /dev/null
@@ -1,75 +0,0 @@
-/*
- *  Copyright 2019 The TensorFlow Authors. All Rights Reserved.
- *
- *  Licensed under the Apache License, Version 2.0 (the "License");
- *  you may not use this file except in compliance with the License.
- *  You may obtain a copy of the License at
- *
- *      http://www.apache.org/licenses/LICENSE-2.0
- *
- *  Unless required by applicable law or agreed to in writing, software
- *  distributed under the License is distributed on an "AS IS" BASIS,
- *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- *  See the License for the specific language governing permissions and
- *  limitations under the License.
- *  =======================================================================
- */
-
-package org.tensorflow;
-
-import java.util.HashMap;
-import java.util.Map;
-import org.tensorflow.types.TBfloat16;
-import org.tensorflow.types.TBool;
-import org.tensorflow.types.TFloat16;
-import org.tensorflow.types.TFloat32;
-import org.tensorflow.types.TFloat64;
-import org.tensorflow.types.TInt32;
-import org.tensorflow.types.TInt64;
-import org.tensorflow.types.TString;
-import org.tensorflow.types.TUint8;
-
-/**
- * Utility class for working with {@link DataType} objects.
- */
-final class DataTypes {
-
-  /**
-   * Find a data type from the type code returned by the native layer (C API).
-   *
-   * <p>Only data types registered via {@link #register(DataType)} can be resolved.
-   *
-   * @param nativeCode native code
-   * @return data type for this code
-   * @throws IllegalArgumentException if the code matches no registered data type
-   */
-  static DataType<?> fromNativeCode(int nativeCode) {
-    DataType<?> dataType = DATA_TYPE_REGISTRY.get(nativeCode);
-    if (dataType == null) {
-      throw new IllegalArgumentException(
-          "DataType " + nativeCode + " is not recognized in Java (version " + TensorFlow.version() + ")");
-    }
-    return dataType;
-  }
-
-  private static final Map<Integer, DataType<?>> DATA_TYPE_REGISTRY = new HashMap<>();
-
-  static {
-    register(TBool.DTYPE);
-    register(TFloat64.DTYPE);
-    register(TFloat32.DTYPE);
-    register(TFloat16.DTYPE);
-    register(TInt32.DTYPE);
-    register(TInt64.DTYPE);
-    register(TString.DTYPE);
-    register(TUint8.DTYPE);
-    register(TBfloat16.DTYPE);
-  }
-
-  // TODO (karllessard): Right now this method is private but we might want to expose it
-  //      to allow user to register custom data types?
-  private static void register(DataType<?> dataType) {
-    DATA_TYPE_REGISTRY.put(dataType.nativeCode(), dataType);
-    DATA_TYPE_REGISTRY.put(dataType.nativeCode() + 100, dataType);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperation.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperation.java
index 012981ac59c..59da4d92dab 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperation.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperation.java
@@ -29,6 +29,7 @@
 import org.tensorflow.internal.c_api.TF_Status;
 import org.tensorflow.internal.c_api.TF_Tensor;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.proto.framework.DataType;
 
 /**
  * Implementation of an {@link Operation} executed eagerly.
@@ -89,12 +90,6 @@ public TFE_TensorHandle getUnsafeNativeHandle(int outputIndex) {
 
   @Override
   public Shape shape(int outputIndex) {
-    // If the tensor of this output has already been resolved, return its shape.
-    // Otherwise, retrieve the tensor shape from the native library.
-    Tensor<?> tensor = outputTensors.get(outputIndex);
-    if (tensor != null) {
-      return tensor.shape();
-    }
     TFE_TensorHandle outputNativeHandle = getUnsafeNativeHandle(outputIndex);
     long[] shape = new long[numDims(outputNativeHandle)];
     for (int i = 0; i < shape.length; ++i) {
@@ -104,20 +99,14 @@ public Shape shape(int outputIndex) {
   }
 
   @Override
-  public DataType<?> dtype(int outputIndex) {
-    // If the tensor of this output has already been resolved, return its datatype.
-    // Otherwise, retrieve the tensor datatype from the native library.
-    Tensor<?> tensor = outputTensors.get(outputIndex);
-    if (tensor != null) {
-      return tensor.dataType();
-    }
+  public DataType dtype(int outputIndex) {
     TFE_TensorHandle outputNativeHandle = getUnsafeNativeHandle(outputIndex);
-    return DataTypes.fromNativeCode(dataType(outputNativeHandle));
+    return DataType.forNumber(dataType(outputNativeHandle));
   }
 
   @Override
-  public Tensor<?> tensor(int outputIndex) {
-    Tensor<?> tensor = outputTensors.get(outputIndex);
+  public TensorHandle tensor(int outputIndex) {
+    TensorHandle tensor = outputTensors.get(outputIndex);
     if (tensor == null) {
       tensor = resolveTensor(outputIndex);
     }
@@ -127,15 +116,15 @@ public Tensor<?> tensor(int outputIndex) {
   private final EagerSession session;
   private final String type;
   private final String name;
-  private final AtomicReferenceArray<Tensor<?>> outputTensors;
+  private final AtomicReferenceArray<TensorHandle> outputTensors;
 
-  private Tensor<?> resolveTensor(int outputIndex) {
+  private TensorHandle resolveTensor(int outputIndex) {
     // Take an optimistic approach, where we attempt to resolve the output tensor without locking.
     // If another thread has resolved it meanwhile, release our copy and reuse the existing one
     // instead.
-    Tensor<?> tensor = resolveTensorHandle(getUnsafeNativeHandle(outputIndex), session);
+    TensorHandle tensor = resolveTensorHandle(getUnsafeNativeHandle(outputIndex), session);
     if (!outputTensors.compareAndSet(outputIndex, null, tensor)) {
-      session.detach(tensor.nativeHandle());
+      session.detach(tensor.get());
       tensor = outputTensors.get(outputIndex);
     }
     return tensor;
@@ -156,13 +145,15 @@ private static void requireTensorHandle(TFE_TensorHandle handle) {
     }
   }
 
-  private static Tensor<?> resolveTensorHandle(TFE_TensorHandle handle, EagerSession session) {
+  private static TensorHandle resolveTensorHandle(TFE_TensorHandle handle, EagerSession session) {
     requireTensorHandle(handle);
     try (PointerScope scope = new PointerScope()) {
       TF_Status status = TF_Status.newStatus();
-      TF_Tensor tensor = TFE_TensorHandleResolve(handle, status).withDeallocator();
+      TF_Tensor nativeTensor = TFE_TensorHandleResolve(handle, status).withDeallocator();
       status.throwExceptionIfNotOK();
-      return Tensor.fromHandle(tensor, session);
+      TensorHandle t = TensorHandle.of(nativeTensor);
+      t.attachTo(session);
+      return t;
     }
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperationBuilder.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperationBuilder.java
index f14795df55a..c33d945c2d6 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperationBuilder.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/EagerOperationBuilder.java
@@ -48,6 +48,8 @@
 import org.tensorflow.internal.c_api.TF_Status;
 import org.tensorflow.internal.c_api.TF_Tensor;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.types.TypeRegistry;
+import org.tensorflow.types.family.TType;
 
 /**
  * An {@link OperationBuilder} for building {@link Operation Operations} that are executed eagerly.
@@ -159,29 +161,29 @@ public EagerOperationBuilder setAttr(String name, boolean[] values) {
   }
 
   @Override
-  public EagerOperationBuilder setAttr(String name, DataType<?> value) {
-    setAttrType(opHandle, name, value.nativeCode());
+  public EagerOperationBuilder setAttr(String name, Class<? extends TType> type) {
+    setAttrType(opHandle, name, TypeRegistry.find(type).dataType().getNumber());
     return this;
   }
 
   @Override
-  public EagerOperationBuilder setAttr(String name, DataType<?>[] values) {
-    int[] c = new int[values.length];
-    for (int i = 0; i < values.length; ++i) {
-      c[i] = values[i].nativeCode();
+  public EagerOperationBuilder setAttr(String name, Class<? extends TType>[] types) {
+    int[] c = new int[types.length];
+    for (int i = 0; i < types.length; ++i) {
+      c[i] = TypeRegistry.find(types[i]).dataType().getNumber();
     }
     setAttrTypeList(opHandle, name, c);
     return this;
   }
 
   @Override
-  public EagerOperationBuilder setAttr(String name, Tensor<?> value) {
-    setAttrTensor(opHandle, name, value.nativeHandle());
+  public EagerOperationBuilder setAttr(String name, TType value) {
+    setAttrTensor(opHandle, name, value.handle().get());
     return this;
   }
 
   @Override
-  public EagerOperationBuilder setAttr(String name, Tensor<?>[] values) {
+  public EagerOperationBuilder setAttr(String name, TType[] values) {
     // TODO (karllessard) could be supported by adding this attribute type in the eager C API
     throw new UnsupportedOperationException(
         "Tensor list attributes are not supported in eager mode");
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Graph.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Graph.java
index 142d481c04f..8a106983771 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Graph.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Graph.java
@@ -54,6 +54,7 @@
 import org.tensorflow.proto.framework.GraphDef;
 import org.tensorflow.proto.util.SaverDef;
 import org.tensorflow.types.TString;
+import org.tensorflow.types.family.TType;
 
 
 /**
@@ -766,15 +767,15 @@ private static SaverDef addVariableSaver(Graph graph) {
     Ops tf = Ops.create(graph).withSubScope("save");
 
     List<String> varNames = new ArrayList<>();
-    List<Operand<?>> varOutputs = new ArrayList<>();
-    List<DataType<?>> varTypes = new ArrayList<>();
+    List<Operand<? extends TType>> varOutputs = new ArrayList<>();
+    List<Class<? extends TType>> varTypes = new ArrayList<>();
 
     for (Iterator<Operation> iter = graph.operations(); iter.hasNext();) {
       Operation op = iter.next();
       if (op.type().equals("VariableV2")) {
         varNames.add(op.name());
         varOutputs.add(op.output(0));
-        varTypes.add(op.output(0).dataType());
+        varTypes.add(op.output(0).type());
       }
     }
 
@@ -783,7 +784,7 @@ private static SaverDef addVariableSaver(Graph graph) {
     Constant<TString> varNamesTensor = tf.constant(StdArrays.ndCopyOf(varNames.toArray(tmp)));
     Operand<TString> varSlices = tf.zerosLike(varNamesTensor);
 
-    Placeholder<TString> saveFilename = tf.placeholder(TString.DTYPE);
+    Placeholder<TString> saveFilename = tf.placeholder(TString.class);
     Save saveVariables = tf.train.save(
         saveFilename,
         varNamesTensor,
@@ -798,7 +799,7 @@ private static SaverDef addVariableSaver(Graph graph) {
     );
     List<Op> restoreOps = new ArrayList<>(varOutputs.size());
     for (int i = 0; i < varOutputs.size(); ++i) {
-      restoreOps.add(tf.assign(varOutputs.get(i), (Operand) restoreVariables.tensors().get(i)));
+      restoreOps.add(tf.assign(varOutputs.get(i), (Operand)restoreVariables.tensors().get(i)));
     }
     NoOp restoreAll = tf.withControlDependencies(restoreOps).noOp();
 
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperation.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperation.java
index 70cd31366ce..36e57d5325c 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperation.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperation.java
@@ -31,6 +31,8 @@
 import org.tensorflow.internal.c_api.TF_Output;
 import org.tensorflow.internal.c_api.TF_Status;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.family.TType;
 
 /**
  * Implementation for an {@link Operation} added as a node to a {@link Graph}.
@@ -148,17 +150,17 @@ Shape shape(int outputIdx) {
   }
 
   @Override
-  DataType<?> dtype(int outputIdx) {
+  DataType dtype(int outputIdx) {
     Graph.Reference r = graph.ref();
     try {
-      return DataTypes.fromNativeCode(dtype(r.nativeHandle(), getUnsafeNativeHandle(), outputIdx));
+      return DataType.forNumber(dtype(r.nativeHandle(), getUnsafeNativeHandle(), outputIdx));
     } finally {
       r.close();
     }
   }
 
   @Override
-  Tensor<?> tensor(int outputIdx) {
+  TensorHandle tensor(int outputIdx) {
     throw new IllegalStateException("Graph tensors must be fetched by running a session");
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperationBuilder.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperationBuilder.java
index 2ef5c9010a1..a10ce0af010 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperationBuilder.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/GraphOperationBuilder.java
@@ -52,6 +52,8 @@
 import org.tensorflow.internal.c_api.TF_Status;
 import org.tensorflow.internal.c_api.TF_Tensor;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.types.TypeRegistry;
+import org.tensorflow.types.family.TType;
 
 /** An {@link OperationBuilder} for adding {@link GraphOperation}s to a {@link Graph}. */
 public final class GraphOperationBuilder implements OperationBuilder {
@@ -221,10 +223,10 @@ public GraphOperationBuilder setAttr(String name, boolean[] value) {
   }
 
   @Override
-  public GraphOperationBuilder setAttr(String name, DataType value) {
+  public GraphOperationBuilder setAttr(String name, Class<? extends TType> type) {
     Graph.Reference r = graph.ref();
     try {
-      setAttrType(unsafeNativeHandle, name, value.nativeCode());
+      setAttrType(unsafeNativeHandle, name, TypeRegistry.find(type).dataType().getNumber());
     } finally {
       r.close();
     }
@@ -232,10 +234,10 @@ public GraphOperationBuilder setAttr(String name, DataType value) {
   }
 
   @Override
-  public GraphOperationBuilder setAttr(String name, DataType[] value) {
-    int[] ctypes = new int[value.length];
-    for (int i = 0; i < value.length; ++i) {
-      ctypes[i] = value[i].nativeCode();
+  public GraphOperationBuilder setAttr(String name, Class<? extends TType>[] types) {
+    int[] ctypes = new int[types.length];
+    for (int i = 0; i < types.length; ++i) {
+      ctypes[i] = TypeRegistry.find(types[i]).dataType().getNumber();
     }
     Graph.Reference r = graph.ref();
     try {
@@ -247,10 +249,10 @@ public GraphOperationBuilder setAttr(String name, DataType[] value) {
   }
 
   @Override
-  public GraphOperationBuilder setAttr(String name, Tensor<?> value) {
+  public GraphOperationBuilder setAttr(String name, TType value) {
     Graph.Reference r = graph.ref();
     try {
-      setAttrTensor(unsafeNativeHandle, name, value.nativeHandle());
+      setAttrTensor(unsafeNativeHandle, name, value.handle().get());
     } finally {
       r.close();
     }
@@ -258,11 +260,11 @@ public GraphOperationBuilder setAttr(String name, Tensor<?> value) {
   }
 
   @Override
-  public GraphOperationBuilder setAttr(String name, Tensor<?>[] value) {
+  public GraphOperationBuilder setAttr(String name, TType[] value) {
     TF_Tensor[] handles = new TF_Tensor[value.length];
     int idx = 0;
-    for (Tensor<?> t : value) {
-      handles[idx++] = t.nativeHandle();
+    for (TType t : value) {
+      handles[idx++] = t.handle().get();
     }
     Graph.Reference r = graph.ref();
     try {
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Operand.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Operand.java
index fa21f32d4ce..fcb630484d0 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Operand.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Operand.java
@@ -15,7 +15,10 @@
 
 package org.tensorflow;
 
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.Op;
+import org.tensorflow.op.Scope;
+import org.tensorflow.types.TypeRegistry;
 import org.tensorflow.types.family.TType;
 
 /**
@@ -41,6 +44,17 @@
  */
 public interface Operand<T extends TType> extends Op {
 
+  /** Returns the (possibly partially known) shape of the tensor referred to by this Output. */
+  default Shape shape() {
+    return asOutput().shape();
+  }
+
+  /** Returns the DataType of the tensor referred to by this Output. */
+  default Class<T> type() {
+    return asOutput().type();
+  }
+
+
   /**
    * Returns the symbolic handle of the tensor.
    *
@@ -60,20 +74,15 @@ public interface Operand<T extends TType> extends Op {
    * @return the tensor
    * @throws IllegalStateException if this is an operand of a graph
    */
-  default Tensor<T> asTensor() {
+  default T asTensor() {
     return asOutput().tensor();
   }
 
-  /**
-   * Returns the data of this operand.
-   *
-   * <i>Only works when running in an eager execution</i>
-   * <p>This helper method is equivalent to {@code asTensor().data()}
-   *
-   * @return the tensor data
-   * @throws IllegalStateException if this is an operand of a graph
-   */
-  default T data() {
-    return asOutput().tensor().data();
+  default <U extends TType> Operand<U> expect(Class<U> type) {
+    if (asOutput().type() != type) {
+      throw new IllegalArgumentException(
+          "Cannot cast from tensor of " + asOutput().type().getSimpleName() + " to tensor of " + type.getSimpleName());
+    }
+    return (Operand<U>)this;
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/OperationBuilder.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/OperationBuilder.java
index af1b8cc9130..df6a80a79ca 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/OperationBuilder.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/OperationBuilder.java
@@ -16,6 +16,7 @@
 package org.tensorflow;
 
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.types.family.TType;
 
 /**
  * A builder for {@link Operation}s.
@@ -177,7 +178,7 @@ public interface OperationBuilder {
    * @param value attribute value
    * @return the OperationBuilder instance for chaining.
    */
-   OperationBuilder setAttr(String name, DataType<?> value);
+   OperationBuilder setAttr(String name, Class<? extends TType> value);
 
   /**
    * Set the type values of an attribute of the operation being built.
@@ -186,7 +187,7 @@ public interface OperationBuilder {
    * @param value attribute values
    * @return the OperationBuilder instance for chaining.
    */
-   OperationBuilder setAttr(String name, DataType<?>[] value);
+   OperationBuilder setAttr(String name, Class<? extends TType>[] value);
 
   /**
    * Set the tensor value of an attribute of the operation being built.
@@ -195,7 +196,7 @@ public interface OperationBuilder {
    * @param value attribute value
    * @return the OperationBuilder instance for chaining.
    */
-   OperationBuilder setAttr(String name, Tensor<?> value);
+   OperationBuilder setAttr(String name, TType value);
 
   /**
    * Set the tensor values of an attribute of the operation being built.
@@ -204,7 +205,7 @@ public interface OperationBuilder {
    * @param value attribute values
    * @return the OperationBuilder instance for chaining.
    */
-   OperationBuilder setAttr(String name, Tensor<?>[] value);
+   OperationBuilder setAttr(String name, TType[] value);
 
   /**
    * Set the shape value of an attribute of the operation being built.
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Output.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Output.java
index a873df8ff4c..7397d676393 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Output.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Output.java
@@ -18,6 +18,9 @@
 import java.util.Objects;
 import org.bytedeco.javacpp.Pointer;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.op.Scope;
+import org.tensorflow.types.Type;
+import org.tensorflow.types.TypeRegistry;
 import org.tensorflow.types.family.TType;
 
 /**
@@ -37,29 +40,30 @@ public int index() {
   }
 
   /** Returns the (possibly partially known) shape of the tensor referred to by this Output. */
+  @Override
   public Shape shape() {
     return operation.shape(index);
   }
 
   /** Returns the DataType of the tensor referred to by this Output. */
-  @SuppressWarnings("unchecked")
-  public DataType<T> dataType() {
-    return (DataType<T>)operation.dtype(index);
+  @Override
+  public Class<T> type() {
+    return ((Type<T>)TypeRegistry.find(operation.dtype(index))).typeClass();
   }
 
   /**
    * Returns this Output object with the type {@code Output<U>}. This method is useful when given a
    * value of type {@code Output<?>}.
    *
-   * @param dt any supported tensor data type
+   * @param tensorType type of tensor at this output
    * @throws IllegalArgumentException if the actual data type of this object does not match the type
    *     {@code U}.
    */
   @SuppressWarnings("unchecked")
-  public <U extends TType> Output<U> expect(DataType<U> dt) {
-    if (!dt.equals(this.dataType())) {
+  public <U extends TType> Output<U> expect(Class<U> tensorType) {
+    if (tensorType != type()) {
       throw new IllegalArgumentException(
-          "Cannot cast from output of " + this.dataType() + " to output of " + dt);
+          "Cannot cast from output of " + type().getSimpleName() + " to output of type " + tensorType.getSimpleName());
     }
     return ((Output<U>) this);
   }
@@ -80,8 +84,8 @@ public <U extends TType> Output<U> expect(DataType<U> dt) {
    * @see EagerSession
    */
   @SuppressWarnings("unchecked")
-  public Tensor<T> tensor() {
-    return (Tensor<T>) operation.tensor(index);
+  public T tensor() {
+    return Tensors.fromHandle(operation.tensor(index));
   }
 
   @Override
@@ -115,7 +119,7 @@ public boolean equals(Object o) {
   public String toString() {
     return String.format(
         "<%s '%s:%d' shape=%s dtype=%s>",
-        operation.type(), operation.name(), index, shape().toString(), dataType());
+        operation.type(), operation.name(), index, shape().toString(), type().getSimpleName());
   }
 
   /** Handle to the idx-th output of the Operation {@code op}. */
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/SavedModelBundle.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/SavedModelBundle.java
index 093898ae56c..6a5ae0b6467 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/SavedModelBundle.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/SavedModelBundle.java
@@ -46,6 +46,8 @@
 import org.tensorflow.proto.framework.RunOptions;
 import org.tensorflow.proto.framework.SavedModel;
 import org.tensorflow.proto.util.SaverDef;
+import org.tensorflow.types.family.TType;
+import org.tensorflow.util.TensorMap;
 
 /**
  * SavedModelBundle represents a model loaded from storage.
@@ -334,7 +336,7 @@ public ConcreteFunction function(String signatureKey) {
    * @return list of output tensors, mapped by the signature name
    * @throws IllegalArgumentException if no function can be selected by default
    */
-  public Map<String, Tensor<?>> call(Map<String, Tensor<?>> arguments) {
+  public TensorMap call(Map<String, TType> arguments) {
     ConcreteFunction function = null;
     if (functions.size() == 1) {
       function = functions.values().iterator().next();
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Session.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Session.java
index 4e82f3944b8..07e70db6dc9 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Session.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Session.java
@@ -15,7 +15,15 @@
 
 package org.tensorflow;
 
+import static org.tensorflow.Graph.resolveOutputs;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_CloseSession;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_DeleteSession;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_NewSession;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_SessionRun;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_SetConfig;
+
 import com.google.protobuf.InvalidProtocolBufferException;
+import java.util.ArrayList;
 import org.bytedeco.javacpp.BytePointer;
 import org.bytedeco.javacpp.Pointer;
 import org.bytedeco.javacpp.PointerPointer;
@@ -33,15 +41,11 @@
 import org.tensorflow.proto.framework.ConfigProto;
 import org.tensorflow.proto.framework.RunMetadata;
 import org.tensorflow.proto.framework.RunOptions;
-
-import java.util.ArrayList;
-import java.util.List;
 import org.tensorflow.proto.util.SaverDef;
+import org.tensorflow.types.family.TType;
+import org.tensorflow.util.TensorList;
 import org.tensorflow.types.TString;
 
-import static org.tensorflow.Graph.resolveOutputs;
-import static org.tensorflow.internal.c_api.global.tensorflow.*;
-
 /**
  * Driver for {@link Graph} execution.
  *
@@ -158,7 +162,7 @@ public final class Runner {
      * @param t the tensor substituting the operation
      * @return this session runner
      */
-    public Runner feed(String operation, Tensor<?> t) {
+    public Runner feed(String operation, TType t) {
       return feed(parseOutput(operation), t);
     }
 
@@ -173,11 +177,11 @@ public Runner feed(String operation, Tensor<?> t) {
      * @param t the tensor substituting the operation
      * @return this session runner
      */
-    public Runner feed(String operation, int index, Tensor<?> t) {
+    public Runner feed(String operation, int index, TType t) {
       Operation op = operationByName(operation);
       if (op != null) {
         inputs.add(op.output(index));
-        inputTensors.add(t);
+        inputTensors.add(t.handle());
       }
       return this;
     }
@@ -190,9 +194,9 @@ public Runner feed(String operation, int index, Tensor<?> t) {
      * @param t the tensor substituting the operation
      * @return this session runner
      */
-    public Runner feed(Operand<?> operand, Tensor<?> t) {
+    public Runner feed(Operand<?> operand, TType t) {
       inputs.add(operand.asOutput());
-      inputTensors.add(t);
+      inputTensors.add(t.handle());
       return this;
     }
 
@@ -325,7 +329,7 @@ public Runner setOptions(RunOptions options) {
      *
      * @return list of resulting tensors fetched by this session runner
      */
-    public List<Tensor<?>> run() {
+    public TensorList run() {
       return runHelper(false).outputs;
     }
 
@@ -354,8 +358,8 @@ private Run runHelper(boolean wantMetadata) {
       // It's okay to use Operation.getUnsafeNativeHandle() here since the safety depends on the
       // validity of the Graph and graphRef ensures that.
       int idx = 0;
-      for (Tensor<?> t : inputTensors) {
-        inputTensorHandles[idx++] = t.nativeHandle();
+      for (TensorHandle t : inputTensors) {
+        inputTensorHandles[idx++] = t.get();
       }
       idx = 0;
       for (Output<?> o : inputs) {
@@ -375,7 +379,7 @@ private Run runHelper(boolean wantMetadata) {
       }
       Reference runRef = new Reference();
       RunMetadata metadata = null;
-      List<Tensor<?>> outputs = new ArrayList<>();
+      TensorList outputs = new TensorList();
       try {
         metadata =
             Session.run(
@@ -390,10 +394,7 @@ private Run runHelper(boolean wantMetadata) {
                 wantMetadata,
                 outputs);
       } catch (Exception e) {
-        for (Tensor<?> t : outputs) {
-          t.close();
-        }
-        outputs.clear();
+        outputs.close();
         throw e;
       } finally {
         runRef.close();
@@ -451,7 +452,7 @@ private Output<?> parseOutput(String opName) {
     }
 
     private ArrayList<Output<?>> inputs = new ArrayList<>();
-    private ArrayList<Tensor<?>> inputTensors = new ArrayList<>();
+    private ArrayList<TensorHandle> inputTensors = new ArrayList<>();
     private ArrayList<Output<?>> outputs = new ArrayList<>();
     private ArrayList<GraphOperation> targets = new ArrayList<>();
     private RunOptions runOptions = null;
@@ -518,7 +519,7 @@ public void save(String prefix) {
    */
   public static final class Run {
     /** Tensors from requested fetches. */
-    public List<Tensor<?>> outputs;
+    public TensorList outputs;
 
     /**
      * Metadata about the run.
@@ -627,7 +628,7 @@ private static RunMetadata run(
       int[] outputOpIndices,
       TF_Operation[] targetOpHandles,
       boolean wantRunMetadata,
-      List<Tensor<?>> outputTensors) {
+      TensorList outputTensors) {
     requireHandle(handle);
 
     int ninputs = inputTensorHandles.length;
@@ -667,7 +668,7 @@ private static RunMetadata run(
 
       for (int i = 0; i < noutputs; ++i) {
         TF_Tensor h = outputValues.get(TF_Tensor.class, i).withDeallocator();
-        outputTensors.add(Tensor.fromHandle(h));
+        outputTensors.add(Tensors.fromHandle(TensorHandle.of(h)));
       }
       try {
         return runMetadata != null ? RunMetadata.parseFrom(runMetadata.dataAsByteBuffer()) : null;
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Signature.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Signature.java
index 376dc9039fc..75785ece8ec 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Signature.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Signature.java
@@ -18,11 +18,11 @@
 import java.util.Map;
 import java.util.Set;
 import org.tensorflow.ndarray.Shape;
-import org.tensorflow.proto.framework.DataType;
 import org.tensorflow.proto.framework.SignatureDef;
 import org.tensorflow.proto.framework.TensorInfo;
 import org.tensorflow.proto.framework.TensorShapeProto;
 import org.tensorflow.proto.framework.TensorShapeProto.Dim;
+import org.tensorflow.types.TypeRegistry;
 
 /**
  * Describe the inputs and outputs of an executable entity, such as a {@link ConcreteFunction}, among
@@ -113,7 +113,7 @@ private static TensorInfo toTensorInfo(Output<?> operand) {
         tensorShapeBuilder.addDim(Dim.newBuilder().setSize(shape.size(i)));
       }
       return TensorInfo.newBuilder()
-          .setDtype(DataType.forNumber(operand.dataType().nativeCode()))
+          .setDtype(TypeRegistry.find(operand.type()).dataType())
           .setTensorShape(tensorShapeBuilder)
           .setName(operand.op().name() + ":" + operand.index())
           .build();
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Tensor.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Tensor.java
index 6787713418f..4f47bd06dd2 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Tensor.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Tensor.java
@@ -15,19 +15,9 @@
 
 package org.tensorflow;
 
-import static org.tensorflow.internal.c_api.global.tensorflow.TF_Dim;
-import static org.tensorflow.internal.c_api.global.tensorflow.TF_NumDims;
-import static org.tensorflow.internal.c_api.global.tensorflow.TF_TensorByteSize;
-import static org.tensorflow.internal.c_api.global.tensorflow.TF_TensorType;
-
-import java.util.function.Consumer;
-import org.bytedeco.javacpp.PointerScope;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.NdArray;
-import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.NdArrayBase;
 import org.tensorflow.ndarray.buffer.ByteDataBuffer;
-import org.tensorflow.types.family.TType;
+import org.tensorflow.proto.framework.DataType;
 
 /**
  * A statically typed multi-dimensional array whose elements are of a type described by T.
@@ -44,167 +34,12 @@
  * }
  * }</pre>
  */
-public final class Tensor<T extends TType> implements AutoCloseable {
-
-  /**
-   * Allocates a tensor of a given datatype and shape.
-   *
-   * <p>The amount of memory to allocate is derived from the datatype and the shape of the tensor.
-   * Memory is left uninitialized after this method returns, so it is the responsibility of the
-   * caller to initialize the tensor data before it is used, via the {@link #data()} accessor.
-   * For example:
-   *
-   * <pre>{@code
-   * FloatNdArray data = ...
-   * try (Tensor<TFloat32> t = Tensor.of(TFloat32.DTYPE, Shape.of(2, 2))) {
-   *   data.copyTo(t.data());
-   *   ...
-   * }
-   * }</pre>
-   *
-   * @param <T> the tensor element type
-   * @param dtype datatype of the tensor
-   * @param shape shape of the tensor
-   * @return an allocated but uninitialized tensor
-   * @throws IllegalStateException if tensor failed to be allocated
-   */
-  public static <T extends TType> Tensor<T> of(DataType<T> dtype, Shape shape) {
-    return of(dtype, shape, shape.size() * dtype.byteSize());
-  }
-
-  /**
-   * Allocates a tensor of a given datatype, shape and size.
-   *
-   * <p>This method is identical to {@link #of(DataType, Shape)}, except that the final size of the
-   * tensor is explicitly set instead of computing it from the datatype and shape.
-   *
-   * <p>This could be useful for tensor types that stores data but also metadata in the tensor memory,
-   * like {@link org.tensorflow.types.TString TString}.
-   *
-   * @param <T> the tensor element type
-   * @param dtype datatype of the tensor
-   * @param shape shape of the tensor
-   * @param size size, in bytes, of the tensor
-   * @return an allocated but uninitialized tensor
-   * @see #of(DataType, Shape)
-   * @throws IllegalArgumentException if {@code size} is smaller than the minimum space required to
-   *                                  store the tensor data
-   * @throws IllegalStateException if tensor failed to be allocated
-   */
-  public static <T extends TType> Tensor<T> of(DataType<T> dtype, Shape shape, long size) {
-    // Minimum requirements for datatypes of variable length cannot be verified in a relevant way so
-    // we only validate them for fixed length datatypes
-    if (!dtype.isVariableLength() && shape.size() * dtype.byteSize() > size) {
-      throw new IllegalArgumentException("Tensor size is not large enough to contain all scalar values");
-    }
-    Tensor<T> t = new Tensor<>(dtype, shape);
-    TF_Tensor nativeHandle = allocate(t.dtype.nativeCode(), shape.asArray(), size);
-    try (PointerScope scope = new PointerScope()) {
-        scope.attach(nativeHandle);
-        t.tensorHandle = nativeHandle;
-        t.tensorScope = scope.extend();
-        return t;
-    }
-  }
-
-  /**
-   * Allocates and initialize a tensor of a given datatype and shape.
-   *
-   * <p>The amount of memory to allocate is derived from the datatype and the shape of the tensor.
-   * Tensor data is initialized by calling the {@code dataInitializer}, which receives in argument
-   * the value returned by {@link #data()} on the allocated tensor. For example:
-   *
-   * <pre>{@code
-   * FloatNdArray data = ...
-   * try (Tensor<TFloat32> t = Tensor.of(TFloat32.DTYPE, Shape.of(2, 2), data::copyTo)) {
-   *   ...
-   * }
-   * }</pre>
-   *
-   * <p>If {@code dataInitializer} fails and throws an exception, the allocated tensor will be
-   * automatically released before rethrowing the same exception.
-   *
-   * @param <T> the tensor element type
-   * @param dtype datatype of the tensor
-   * @param shape shape of the tensor
-   * @param dataInitializer method receiving accessor to the allocated tensor data for initialization
-   * @return an allocated and initialized tensor
-   * @throws IllegalStateException if tensor failed to be allocated
-   */
-  public static <T extends TType> Tensor<T> of(DataType<T> dtype, Shape shape,
-      Consumer<T> dataInitializer) {
-    return of(dtype, shape, shape.size() * dtype.byteSize(), dataInitializer);
-  }
+public interface Tensor extends NdArrayBase, AutoCloseable {
 
   /**
-   * Allocates a tensor of a given datatype, shape and size.
-   *
-   * <p>This method is identical to {@link #of(DataType, Shape, Consumer)}, except that the final
-   * size for the tensor is explicitly set instead of being computed from the datatype and shape.
-   *
-   * <p>This could be useful for tensor types that stores data but also metadata in the tensor memory,
-   * such as {@link org.tensorflow.types.TString TString}.
-   *
-   * @param <T> the tensor element type
-   * @param dtype datatype of the tensor
-   * @param shape shape of the tensor
-   * @param size size, in bytes, of the tensor
-   * @param dataInitializer method receiving accessor to the allocated tensor data for initialization
-   * @return an allocated and initialized tensor
-   * @see #of(DataType, Shape, long, Consumer)
-   * @throws IllegalArgumentException if {@code size} is smaller than the minimum space required to
-   *                                  store the tensor data
-   * @throws IllegalStateException if tensor failed to be allocated
+   * Return the handle to the native tensor
    */
-  public static <T extends TType> Tensor<T> of(DataType<T> dtype, Shape shape, long size,
-      Consumer<T> dataInitializer) {
-    Tensor<T> tensor = of(dtype, shape, size);
-    try {
-      dataInitializer.accept(tensor.data());
-      return tensor;
-    } catch (Throwable t) {
-      tensor.close();
-      throw t;
-    }
-  }
-
-  /**
-   * Creates a Tensor of any type from the raw data provided by the given buffer.
-   *
-   * <p>Data must have been encoded into {@code data} as per the specification of the TensorFlow <a
-   * href="https://www.tensorflow.org/code/tensorflow/c/c_api.h">C API</a>.
-   *
-   * @param <T> the tensor element type
-   * @param dtype the tensor element data type
-   * @param shape the tensor shape.
-   * @param rawData a buffer containing the tensor raw data.
-   * @throws IllegalArgumentException if {@code rawData} is not large enough to contain the tensor data
-   * @throws IllegalStateException if tensor failed to be allocated with the given parameters
-   */
-  public static <T extends TType> Tensor<T> of(DataType<T> dtype, Shape shape, ByteDataBuffer rawData) {
-    Tensor<T> t = of(dtype, shape, rawData.size());
-    rawData.copyTo(TensorBuffers.toBytes(t.nativeHandle()), rawData.size());
-    return t;
-  }
-
-  /**
-   * Returns this Tensor object with the type {@code Tensor<U>}. This method is useful when given a
-   * value of type {@code Tensor<?>}.
-   *
-   * @param dt any supported tensor data type
-   * @param <U> a tensor type
-   * @return a tensor of the requested data type
-   * @throws IllegalArgumentException if the actual data type of this object does not match the type
-   *     {@code U}.
-   */
-  @SuppressWarnings("unchecked")
-  public <U extends TType> Tensor<U> expect(DataType<U> dt) {
-    if (!dt.equals(this.dtype)) {
-      throw new IllegalArgumentException(
-          "Cannot cast from tensor of " + dtype + " to tensor of " + dt);
-    }
-    return ((Tensor<U>) this);
-  }
+  TensorHandle handle();
 
   /**
    * Release resources associated with the Tensor.
@@ -215,73 +50,7 @@ public <U extends TType> Tensor<U> expect(DataType<U> dt) {
    * <p>The Tensor object is no longer usable after {@code close} returns.
    */
   @Override
-  public void close() {
-    tensorScope.close();
-  }
-
-  /** Returns the {@link DataType} of elements stored in the Tensor. */
-  public DataType<T> dataType() {
-    return dtype;
-  }
-
-  /** Returns the size, in bytes, of the tensor data. */
-  public long numBytes() {
-    if (numBytes == null) {
-      numBytes = TF_TensorByteSize(tensorHandle);
-    }
-    return numBytes;
-  }
-
-  /**
-   * Returns the <a href="https://www.tensorflow.org/resources/dims_types.html#shape">shape</a> of
-   * the Tensor, i.e., the sizes of each dimension.
-   *
-   * @return shape of this tensor
-   */
-  public Shape shape() {
-    return shape;
-  }
-
-  /**
-   * Returns the data of this tensor.
-   *
-   * <p>This method returns an accessor to the tensor data as an instance of {@code T}, which
-   * commonly maps this data to an {@link NdArray NdArray}. Input and
-   * output operations performed on the returned n-dimensional array are applied directly to the
-   * tensor native memory. For example:
-   *
-   * <pre>{@code
-   * Ops tf = Ops.create();
-   * try (Tensor<TFloat32> t = TFloat32.tensorOf(Shape.of(2, 2))) {
-   *   TFloat32 data = t.data();
-   *
-   *   StdArrays.copyTo(data, new float[][] {
-   *     {1.0f, 2.0f},
-   *     {3.0f, 4.0f}
-   *   });
-   *   assertEquals(NdArrays.vectorOf(3.0f, 4.0f), data.getFloat(1));
-   *
-   *   Constant<TFloat32> c = tf.constant(t);
-   *   assertEquals(4.0f, c.data().getFloat(1, 1));
-   * }
-   * }</pre>
-   *
-   * <p>Please refer to the documentation of the {@link NdArray NdArray}
-   * classes for more information on the various techniques to read or write data in an
-   * n-dimensional space using this data structure.
-   *
-   * @return the tensor data mapped to an n-dimensional space
-   * @throws IllegalStateException if the tensor has been closed
-   * @see NdArray
-   */
-  public T data() {
-    if (data == null) {
-      data = dtype.map(this);
-    } else {
-      nativeHandle(); // Checks that the tensor has not been released or will throw
-    }
-    return data;
-  }
+  void close();
 
   /**
    * Returns the raw data of this tensor as a buffer of bytes.
@@ -293,95 +62,13 @@ public T data() {
    * @return the tensor raw data mapped to a read-only byte buffer
    * @throws IllegalStateException if the tensor has been closed
    */
-  public ByteDataBuffer rawData() {
-    return TensorBuffers.toBytes(nativeHandle(), true);
-  }
-
-  /** Returns a string describing the type and shape of the Tensor. */
-  @Override
-  public String toString() {
-    return String.format("%s tensor with shape %s", dtype.toString(), shape);
-  }
-
-  /**
-   * Create a Tensor object from a handle to the C TF_Tensor object.
-   *
-   * <p>Takes ownership of the handle.
-   */
-  static Tensor<?> fromHandle(TF_Tensor handle) {
-    Tensor<?> t = new Tensor<>(DataTypes.fromNativeCode(dtype(handle)), Shape.of(shape(handle)));
-    try (PointerScope scope = new PointerScope()) {
-        scope.attach(handle);
-        t.tensorHandle = handle;
-        t.tensorScope = scope.extend();
-    }
-    return t;
-  }
+  ByteDataBuffer rawData();
 
-  /**
-   * Create an eager Tensor object from a handle to the C TF_Tensor object.
-   *
-   * <p>Takes ownership of the handle.
-   */
-  static Tensor<?> fromHandle(TF_Tensor handle, EagerSession session) {
-    Tensor<?> t = fromHandle(handle);
-    session.attach(handle);
-    t.tensorScope.detach(handle);
-    return t;
-  }
-
-  /**
-   * @return native handle to this tensor
-   * @throws IllegalStateException if tensor has been closed
-   */
-  TF_Tensor nativeHandle() {
-    return requireHandle(tensorHandle);
-  }
-
-  private PointerScope tensorScope;
-  private TF_Tensor tensorHandle;
-
-  private static TF_Tensor requireHandle(TF_Tensor handle) {
-    if (handle == null || handle.isNull()) {
-      throw new IllegalStateException("close() was called on the Tensor");
-    }
-    return handle;
-  }
-
-  private static TF_Tensor allocate(int dtype, long[] shape, long byteSize) {
-    TF_Tensor t = TF_Tensor.allocateTensor(dtype, shape, byteSize);
-    if (t == null || t.isNull()) {
-      throw new IllegalStateException("unable to allocate memory for the Tensor");
-    }
-    return t;
-  }
-
-  private static int dtype(TF_Tensor handle) {
-    requireHandle(handle);
-    return TF_TensorType(handle);
-  }
+  /** Returns the size, in bytes, of the tensor data. */
+  long numBytes();
 
-  private static long[] shape(TF_Tensor handle) {
-    requireHandle(handle);
-    int numDims = TF_NumDims(handle);
-    long[] dims = new long[numDims];
-    for (int i = 0; i < numDims; ++i) {
-      dims[i] = TF_Dim(handle, i);
-    }
-    return dims;
-  }
+  DataType dataType();
+}
 
-  private final DataType<T> dtype;
-  private final Shape shape;
-  private T data = null;
-  private Long numBytes = null;
 
-  private Tensor(DataType<T> dtype, Shape shape) {
-    this.dtype = dtype;
-    this.shape = shape;
-  }
 
-  static {
-    TensorFlow.init();
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/TensorHandle.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/TensorHandle.java
new file mode 100644
index 00000000000..7ac9ac291a8
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/TensorHandle.java
@@ -0,0 +1,132 @@
+/* Copyright 2016 The TensorFlow Authors. All Rights Reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+==============================================================================*/
+
+package org.tensorflow;
+
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_Dim;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_NumDims;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_TensorByteSize;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_TensorElementCount;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_TensorType;
+
+import org.tensorflow.internal.c_api.TF_Tensor;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.ByteDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+
+public final class TensorHandle implements Tensor {
+
+  @Override
+  public TensorHandle handle() {
+    return this;
+  }
+
+  public void retain() {
+    requireHandle(nativeHandle).retainReference();
+  }
+
+  public void release() {
+    requireHandle(nativeHandle).releaseReference();
+  }
+
+  @Override
+  public Shape shape() {
+    return Shape.of(shape(nativeHandle));
+  }
+
+  @Override
+  public int rank() {
+    return numDims(nativeHandle);
+  }
+
+  @Override
+  public long size() {
+    return numElements(nativeHandle);
+  }
+
+  @Override
+  public void close() {
+    get().close();
+    this.nativeHandle = null;
+  }
+
+  @Override
+  public long numBytes() {
+    return TF_TensorByteSize(get());
+  }
+
+  @Override
+  public ByteDataBuffer rawData() {
+    return TensorBuffers.toBytes(get(), true);
+  }
+
+  @Override
+  public DataType dataType() {
+    return DataType.forNumber(dtype(get()));
+  }
+
+  static TensorHandle of(TF_Tensor nativeHandle) {
+    return new TensorHandle(nativeHandle);
+  }
+
+  void attachTo(EagerSession session) {
+    session.attach(get());
+    nativeHandle.releaseReference();
+  }
+
+  TF_Tensor get() {
+    return requireHandle(nativeHandle);
+  }
+
+  private TensorHandle(TF_Tensor nativeHandle) {
+    this.nativeHandle = nativeHandle;
+    nativeHandle.retainReference();
+  }
+
+  private static TF_Tensor requireHandle(TF_Tensor handle) {
+    if (handle == null || handle.isNull()) {
+      throw new IllegalStateException("close() was called on the Tensor");
+    }
+    return handle;
+  }
+
+  private static int dtype(TF_Tensor handle) {
+    requireHandle(handle);
+    return TF_TensorType(handle);
+  }
+
+  private static int numDims(TF_Tensor handle) {
+    requireHandle(handle);
+    return TF_NumDims(handle);
+  }
+
+  private static long numElements(TF_Tensor handle) {
+    requireHandle(handle);
+    return TF_TensorElementCount(handle);
+  }
+
+  private static long[] shape(TF_Tensor handle) {
+    requireHandle(handle);
+    int numDims = TF_NumDims(handle);
+    long[] dims = new long[numDims];
+    for (int i = 0; i < numDims; ++i) {
+      dims[i] = TF_Dim(handle, i);
+    }
+    return dims;
+  }
+
+  private TF_Tensor nativeHandle;
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Tensors.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Tensors.java
new file mode 100644
index 00000000000..22c6a769e0f
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Tensors.java
@@ -0,0 +1,201 @@
+package org.tensorflow;
+
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_Dim;
+import static org.tensorflow.internal.c_api.global.tensorflow.TF_NumDims;
+
+import java.util.function.Consumer;
+import org.bytedeco.javacpp.PointerScope;
+import org.tensorflow.internal.c_api.TF_Tensor;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.ByteDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.Type;
+import org.tensorflow.types.TypeRegistry;
+import org.tensorflow.types.family.TType;
+
+public final class Tensors {
+
+  /**
+   * Allocates a tensor of a given datatype and shape.
+   *
+   * <p>The amount of memory to allocate is derived from the datatype and the shape of the tensor.
+   * Memory is left uninitialized after this method returns, so it is the responsibility of the
+   * caller to initialize the tensor data before it is used, via the {@link #data()} accessor.
+   * For example:
+   *
+   * <pre>{@code
+   * FloatNdArray data = ...
+   * try (Tensor<TFloat32> t = Tensor.of(TFloat32.DTYPE, Shape.of(2, 2))) {
+   *   data.copyTo(t.data());
+   *   ...
+   * }
+   * }</pre>
+   *
+   * @param <T> the tensor element type
+   * @param type tensor type
+   * @param shape shape of the tensor
+   * @return an allocated but uninitialized tensor
+   * @throws IllegalStateException if tensor failed to be allocated
+   */
+  public static <T extends TType> T of(Class<T> type, Shape shape) {
+    return of(type, shape, -1);
+  }
+
+  /**
+   * Allocates a tensor of a given datatype, shape and size.
+   *
+   * <p>This method is identical to {@link #of(DataType, Shape)}, except that the final size of the
+   * tensor is explicitly set instead of computing it from the datatype and shape.
+   *
+   * <p>This could be useful for tensor types that stores data but also metadata in the tensor memory,
+   * like {@link org.tensorflow.types.TString TString}.
+   *
+   * @param <T> the tensor element type
+   * @param type tensor type
+   * @param shape shape of the tensor
+   * @param size size, in bytes, of the tensor
+   * @return an allocated but uninitialized tensor
+   * @see #of(DataType, Shape)
+   * @throws IllegalArgumentException if {@code size} is smaller than the minimum space required to
+   *                                  store the tensor data
+   * @throws IllegalStateException if tensor failed to be allocated
+   */
+  public static <T extends TType> T of(Class<T> type, Shape shape, long size) {
+    return allocate(type, shape, size);
+  }
+
+  /**
+   * Allocates and initialize a tensor of a given datatype and shape.
+   *
+   * <p>The amount of memory to allocate is derived from the datatype and the shape of the tensor.
+   * Tensor data is initialized by calling the {@code dataInitializer}, which receives in argument
+   * the value returned by {@link #data()} on the allocated tensor. For example:
+   *
+   * <pre>{@code
+   * FloatNdArray data = ...
+   * try (Tensor<TFloat32> t = Tensor.of(TFloat32.DTYPE, Shape.of(2, 2), data::copyTo)) {
+   *   ...
+   * }
+   * }</pre>
+   *
+   * <p>If {@code dataInitializer} fails and throws an exception, the allocated tensor will be
+   * automatically released before rethrowing the same exception.
+   *
+   * @param <T> the tensor element type
+   * @param type tensor type
+   * @param shape shape of the tensor
+   * @param dataInitializer method receiving accessor to the allocated tensor data for initialization
+   * @return an allocated and initialized tensor
+   * @throws IllegalStateException if tensor failed to be allocated
+   */
+  public static <T extends TType> T of(Class<T> type, Shape shape, Consumer<T> dataInitializer) {
+    return of(type, shape, -1, dataInitializer);
+  }
+
+  /**
+   * Allocates a tensor of a given datatype, shape and size.
+   *
+   * <p>This method is identical to {@link #of(DataType, Shape, Consumer)}, except that the final
+   * size for the tensor is explicitly set instead of being computed from the datatype and shape.
+   *
+   * <p>This could be useful for tensor types that stores data but also metadata in the tensor memory,
+   * such as {@link org.tensorflow.types.TString TString}.
+   *
+   * @param <T> the tensor element type
+   * @param type tensor type
+   * @param shape shape of the tensor
+   * @param size size, in bytes, of the tensor
+   * @param dataInitializer method receiving accessor to the allocated tensor data for initialization
+   * @return an allocated and initialized tensor
+   * @see #of(DataType, Shape, long, Consumer)
+   * @throws IllegalArgumentException if {@code size} is smaller than the minimum space required to
+   *                                  store the tensor data
+   * @throws IllegalStateException if tensor failed to be allocated
+   */
+  public static <T extends TType> T of(Class<T> type, Shape shape, long size, Consumer<T> dataInitializer) {
+    T tensor = of(type, shape, size);
+    try {
+      dataInitializer.accept(tensor);
+      return tensor;
+    } catch (Throwable t) {
+      tensor.close();
+      throw t;
+    }
+  }
+
+  /**
+   * Creates a Tensor of any type from the raw data provided by the given buffer.
+   *
+   * <p>Data must have been encoded into {@code data} as per the specification of the TensorFlow <a
+   * href="https://www.tensorflow.org/code/tensorflow/c/c_api.h">C API</a>.
+   *
+   * @param <T> the tensor element type
+   * @param type tensor type
+   * @param dtype the tensor element data type
+   * @param shape the tensor shape.
+   * @param rawData a buffer containing the tensor raw data.
+   * @throws IllegalArgumentException if {@code rawData} is not large enough to contain the tensor data
+   * @throws IllegalStateException if tensor failed to be allocated with the given parameters
+   */
+  public static <T extends TType> T of(Class<T> type, Shape shape, ByteDataBuffer rawData) {
+    T t = of(type, shape, rawData.size());
+    rawData.copyTo(TensorBuffers.toBytes(t.handle().get()), rawData.size());
+    return t;
+  }
+
+  /**
+   * Create a Tensor object from a handle to the C TF_Tensor object.
+   *
+   * <p>Takes ownership of the handle.
+   */
+  static <T extends TType> T fromHandle(TensorHandle handle) {
+    Type<T> type = TypeRegistry.find(handle.dataType());
+    Shape shape = Shape.of(shape(handle.get()));
+    return type.factory().createDense(handle, shape);
+  }
+
+  private static <T extends TType> T allocate(Class<T> typeClass, Shape shape, long size) {
+    Type<T> type = TypeRegistry.find(typeClass);
+    long effectiveSize = size;
+    if (effectiveSize < 0) {
+      // Size of the tensor is by default the sum of the size of all its element
+      effectiveSize = shape.size() * type.byteSize();
+
+    } else if (!type.isVariableLength() && shape.size() * type.byteSize() > effectiveSize) {
+      // Minimum requirements for datatypes of variable length cannot be verified in a relevant way
+      // so we only validate them for fixed length datatypes
+      throw new IllegalArgumentException("Tensor size is not large enough to contain all scalar values");
+    }
+    TF_Tensor nativeHandle = allocate(type.dataType().getNumber(), shape.asArray(), effectiveSize);
+    try (PointerScope scope = new PointerScope()) {
+      scope.attach(nativeHandle);
+      return type.factory().createDense(TensorHandle.of(nativeHandle), shape);
+    }
+  }
+
+  private static TF_Tensor requireHandle(TF_Tensor handle) {
+    if (handle == null || handle.isNull()) {
+      throw new IllegalStateException("close() was called on the Tensor");
+    }
+    return handle;
+  }
+
+  private static TF_Tensor allocate(int dtype, long[] shape, long byteSize) {
+    TF_Tensor t = TF_Tensor.allocateTensor(dtype, shape, byteSize);
+    if (t == null || t.isNull()) {
+      throw new IllegalStateException("unable to allocate memory for the Tensor");
+    }
+    return t;
+  }
+
+  private static long[] shape(TF_Tensor handle) {
+    requireHandle(handle);
+    int numDims = TF_NumDims(handle);
+    long[] dims = new long[numDims];
+    for (int i = 0; i < numDims; ++i) {
+      dims[i] = TF_Dim(handle, i);
+    }
+    return dims;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/StringTensorBuffer.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/ByteSequenceTensorBuffer.java
similarity index 91%
rename from tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/StringTensorBuffer.java
rename to tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/ByteSequenceTensorBuffer.java
index 83cdab33452..bc80b2a164c 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/StringTensorBuffer.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/ByteSequenceTensorBuffer.java
@@ -50,7 +50,7 @@
  * <p>After its data has been initialized, the buffer is read-only as it is not possible to change
  * safely a value without reinitializing the whole data.
  */
-public class StringTensorBuffer extends AbstractDataBuffer<byte[]> {
+public class ByteSequenceTensorBuffer extends AbstractDataBuffer<byte[]> {
 
   /**
    * Computes how many bytes are required to store the given data in a string buffer.
@@ -66,7 +66,7 @@ public static <T> long computeSize(NdArray<T> data, Function<T, byte[]> getBytes
     // reserve space to store length and data of each values
     for (NdArray<T> scalar : data.scalars()) {
       byte[] elementBytes = getBytes.apply(scalar.getObject());
-      size += elementBytes.length + StringTensorBuffer.varintLength(elementBytes.length);
+      size += elementBytes.length + ByteSequenceTensorBuffer.varintLength(elementBytes.length);
     }
     return size;
   }
@@ -129,8 +129,8 @@ public boolean isReadOnly() {
 
   @Override
   public DataBuffer<byte[]> copyTo(DataBuffer<byte[]> dst, long size) {
-    if (size == size() && dst instanceof StringTensorBuffer) {
-      StringTensorBuffer tensorDst = (StringTensorBuffer) dst;
+    if (size == size() && dst instanceof ByteSequenceTensorBuffer) {
+      ByteSequenceTensorBuffer tensorDst = (ByteSequenceTensorBuffer) dst;
       if (offsets.size() != size || data.size() != size) {
         throw new IllegalArgumentException(
             "Cannot copy string tensor data to another tensor of a different size");
@@ -145,20 +145,20 @@ public DataBuffer<byte[]> copyTo(DataBuffer<byte[]> dst, long size) {
 
   @Override
   public DataBuffer<byte[]> offset(long index) {
-    return new StringTensorBuffer(offsets.offset(index), data);
+    return new ByteSequenceTensorBuffer(offsets.offset(index), data);
   }
 
   @Override
   public DataBuffer<byte[]> narrow(long size) {
-    return new StringTensorBuffer(offsets.narrow(size), data);
+    return new ByteSequenceTensorBuffer(offsets.narrow(size), data);
   }
 
   @Override
   public DataBuffer<byte[]> slice(long index, long size) {
-    return new StringTensorBuffer(offsets.slice(index, size), data);
+    return new ByteSequenceTensorBuffer(offsets.slice(index, size), data);
   }
 
-  StringTensorBuffer(LongDataBuffer offsets, ByteDataBuffer data) {
+  ByteSequenceTensorBuffer(LongDataBuffer offsets, ByteDataBuffer data) {
     this.offsets = offsets;
     this.data = data;
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorBuffers.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorBuffers.java
index f29396dd321..415c5ca35ef 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorBuffers.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorBuffers.java
@@ -156,7 +156,7 @@ public static BooleanDataBuffer toBooleans(TF_Tensor nativeTensor) {
    * @param nativeTensor native reference to the tensor
    * @return a string buffer
    */
-  public static StringTensorBuffer toStrings(TF_Tensor nativeTensor, long numElements) {
+  public static ByteSequenceTensorBuffer toStrings(TF_Tensor nativeTensor, long numElements) {
     Pointer tensorMemory = tensorMemory(nativeTensor);
     if (TensorRawDataBufferFactory.canBeUsed()) {
       return TensorRawDataBufferFactory.mapTensorToStrings(tensorMemory, numElements);
@@ -173,7 +173,7 @@ public static StringTensorBuffer toStrings(TF_Tensor nativeTensor, long numEleme
     dataBuffer.position((int)numElements * Long.BYTES);
     ByteDataBuffer data = DataBuffers.of(dataBuffer.slice());
 
-    return new StringTensorBuffer(offsets, data);
+    return new ByteSequenceTensorBuffer(offsets, data);
   }
 
   private static Pointer tensorMemory(TF_Tensor nativeTensor) {
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorRawDataBufferFactory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorRawDataBufferFactory.java
index 1cfb1c9ab9a..dbaf31f1dcc 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorRawDataBufferFactory.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/buffer/TensorRawDataBufferFactory.java
@@ -57,13 +57,13 @@ static BooleanDataBuffer mapTensorToBooleans(Pointer tensorMemory) {
     return mapNativeBooleans(tensorMemory.address(), tensorMemory.capacity(), false);
   }
 
-  static StringTensorBuffer mapTensorToStrings(Pointer tensorMemory, long numElements) {
+  static ByteSequenceTensorBuffer mapTensorToStrings(Pointer tensorMemory, long numElements) {
     long offsetByteSize = numElements * Long.BYTES;
     LongDataBuffer offsets = mapNativeLongs(tensorMemory.address(), offsetByteSize, false);
     ByteDataBuffer data = mapNativeBytes(
         tensorMemory.address() + offsetByteSize,
         tensorMemory.capacity() - offsetByteSize,
         false);
-    return new StringTensorBuffer(offsets, data);
+    return new ByteSequenceTensorBuffer(offsets, data);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TBfloat16Factory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TBfloat16Factory.java
new file mode 100644
index 00000000000..50542dc0dc7
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TBfloat16Factory.java
@@ -0,0 +1,42 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.ndarray.buffer.layout.DataLayouts;
+import org.tensorflow.ndarray.impl.dense.FloatDenseNdArray;
+import org.tensorflow.types.TBfloat16;
+
+/**
+ * Factory of {@link TBfloat16} tensor instances
+ */
+public class TBfloat16Factory extends AbstractTypeFactory<TBfloat16> {
+
+  @Override
+  public TBfloat16 createDense(TensorHandle tensorHandle, Shape shape) {
+    FloatDataBuffer buffer = DataLayouts.BFLOAT16.applyTo(TensorBuffers.toShorts(getNative(tensorHandle)));
+    return new TBfloat16Impl(tensorHandle, buffer, shape);
+  }
+
+  private static final class TBfloat16Impl extends FloatDenseNdArray implements TBfloat16 {
+
+    @Override
+    public Class<TBfloat16> type() {
+      return TBfloat16.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    TBfloat16Impl(TensorHandle tensorHandle, FloatDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TBoolFactory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TBoolFactory.java
new file mode 100644
index 00000000000..1d1552cc291
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TBoolFactory.java
@@ -0,0 +1,41 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.BooleanDataBuffer;
+import org.tensorflow.ndarray.impl.dense.BooleanDenseNdArray;
+import org.tensorflow.types.TBool;
+
+/**
+ * Factory of {@link TBool} tensor instances
+ */
+public class TBoolFactory extends AbstractTypeFactory<TBool> {
+
+  @Override
+  public TBool createDense(TensorHandle tensorHandle, Shape shape) {
+    BooleanDataBuffer buffer = TensorBuffers.toBooleans(getNative(tensorHandle));
+    return new TBoolImpl(tensorHandle, buffer, shape);
+  }
+
+  private static final class TBoolImpl extends BooleanDenseNdArray implements TBool {
+
+    @Override
+    public Class<TBool> type() {
+      return TBool.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    TBoolImpl(TensorHandle tensorHandle, BooleanDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat16Factory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat16Factory.java
new file mode 100644
index 00000000000..73301f6894e
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat16Factory.java
@@ -0,0 +1,42 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.ndarray.buffer.layout.DataLayouts;
+import org.tensorflow.ndarray.impl.dense.FloatDenseNdArray;
+import org.tensorflow.types.TFloat16;
+
+/**
+ * Factory of {@link TFloat16} tensor instances
+ */
+public class TFloat16Factory extends AbstractTypeFactory<TFloat16> {
+
+  @Override
+  public TFloat16 createDense(TensorHandle tensorHandle, Shape shape) {
+    FloatDataBuffer buffer = DataLayouts.FLOAT16.applyTo(TensorBuffers.toShorts(getNative(tensorHandle)));
+    return new TFloat16Impl(tensorHandle, buffer, shape);
+  }
+
+  private static final class TFloat16Impl extends FloatDenseNdArray implements TFloat16 {
+
+    @Override
+    public Class<TFloat16> type() {
+      return TFloat16.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    TFloat16Impl(TensorHandle tensorHandle, FloatDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat32Factory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat32Factory.java
new file mode 100644
index 00000000000..ae3a53f2c8a
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat32Factory.java
@@ -0,0 +1,41 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.ndarray.impl.dense.FloatDenseNdArray;
+import org.tensorflow.types.TFloat32;
+
+/**
+ * Factory of {@link TFloat32} tensor instances
+ */
+public class TFloat32Factory extends AbstractTypeFactory<TFloat32> {
+
+  @Override
+  public TFloat32 createDense(TensorHandle tensorHandle, Shape shape) {
+    FloatDataBuffer buffer = TensorBuffers.toFloats(getNative(tensorHandle));
+    return new TFloat32Impl(tensorHandle, buffer, shape);
+  }
+
+  private static final class TFloat32Impl extends FloatDenseNdArray implements TFloat32 {
+
+    @Override
+    public Class<TFloat32> type() {
+      return TFloat32.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    TFloat32Impl(TensorHandle tensorHandle, FloatDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat64Factory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat64Factory.java
new file mode 100644
index 00000000000..495f743e8cc
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TFloat64Factory.java
@@ -0,0 +1,41 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.DoubleDataBuffer;
+import org.tensorflow.ndarray.impl.dense.DoubleDenseNdArray;
+import org.tensorflow.types.TFloat64;
+
+/**
+ * Factory of {@link TFloat64} tensor instances
+ */
+public class TFloat64Factory extends AbstractTypeFactory<TFloat64> {
+
+  @Override
+  public TFloat64 createDense(TensorHandle tensorHandle, Shape shape) {
+    DoubleDataBuffer buffer = TensorBuffers.toDoubles(getNative(tensorHandle));
+    return new TFloat64Impl(tensorHandle, buffer, shape);
+  }
+
+  private static final class TFloat64Impl extends DoubleDenseNdArray implements TFloat64 {
+
+    @Override
+    public Class<TFloat64> type() {
+      return TFloat64.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    TFloat64Impl(TensorHandle tensorHandle, DoubleDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TInt32Factory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TInt32Factory.java
new file mode 100644
index 00000000000..368f543d743
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TInt32Factory.java
@@ -0,0 +1,41 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.IntDataBuffer;
+import org.tensorflow.ndarray.impl.dense.IntDenseNdArray;
+import org.tensorflow.types.TInt32;
+
+/**
+ * Factory of {@link TInt32} tensor instances
+ */
+public class TInt32Factory extends AbstractTypeFactory<TInt32> {
+
+  @Override
+  public TInt32 createDense(TensorHandle tensorHandle, Shape shape) {
+    IntDataBuffer buffer = TensorBuffers.toInts(getNative(tensorHandle));
+    return new TInt32Impl(tensorHandle, buffer, shape);
+  }
+
+  private static final class TInt32Impl extends IntDenseNdArray implements TInt32 {
+
+    @Override
+    public Class<TInt32> type() {
+      return TInt32.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    TInt32Impl(TensorHandle tensorHandle, IntDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TInt64Factory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TInt64Factory.java
new file mode 100644
index 00000000000..6a6fc0317ec
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TInt64Factory.java
@@ -0,0 +1,41 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.LongDataBuffer;
+import org.tensorflow.ndarray.impl.dense.LongDenseNdArray;
+import org.tensorflow.types.TInt64;
+
+/**
+ * Factory of {@link TInt64} tensor instances
+ */
+public class TInt64Factory extends AbstractTypeFactory<TInt64> {
+
+  @Override
+  public TInt64 createDense(TensorHandle tensorHandle, Shape shape) {
+    LongDataBuffer buffer = TensorBuffers.toLongs(getNative(tensorHandle));
+    return new Int64DenseTensor(tensorHandle, buffer, shape);
+  }
+
+  private static final class Int64DenseTensor extends LongDenseNdArray implements TInt64 {
+
+    @Override
+    public Class<TInt64> type() {
+      return TInt64.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    Int64DenseTensor(TensorHandle tensorHandle, LongDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TStringFactory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TStringFactory.java
new file mode 100644
index 00000000000..562662579fe
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TStringFactory.java
@@ -0,0 +1,78 @@
+package org.tensorflow.internal.types;
+
+import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
+import java.util.function.Function;
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.ByteSequenceTensorBuffer;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.DataBuffer;
+import org.tensorflow.ndarray.buffer.layout.DataLayout;
+import org.tensorflow.ndarray.buffer.layout.DataLayouts;
+import org.tensorflow.ndarray.impl.dense.DenseNdArray;
+import org.tensorflow.types.TString;
+
+/**
+ * Factory of {@link TString} tensor instances
+ */
+public class TStringFactory extends AbstractTypeFactory<TString> {
+
+  @Override
+  public TString createDense(TensorHandle tensorHandle, Shape shape) {
+    ByteSequenceTensorBuffer buffer = TensorBuffers.toStrings(getNative(tensorHandle), shape.size());
+    return new TStringImpl(tensorHandle, buffer, shape);
+  }
+
+  private static final DataLayout<DataBuffer<byte[]>, String> UTF_8_LAYOUT =
+      DataLayouts.ofStrings(StandardCharsets.UTF_8);
+
+  private static final class TStringImpl extends DenseNdArray<String> implements TString {
+
+    @Override
+    public NdArray<byte[]> asBytes() {
+      return NdArrays.wrap(shape(), rawBuffer);
+    }
+
+    @Override
+    public Class<TString> type() {
+      return TString.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    @Override
+    public TString using(Charset charset) {
+      return new TStringImpl(tensorHandle, rawBuffer, shape(), DataLayouts.ofStrings(charset));
+    }
+
+    @Override
+    public <T> void write(NdArray<T> src, Function<T, byte[]> getBytes) {
+      rawBuffer.init(src, getBytes);
+    }
+
+    TStringImpl(TensorHandle tensorHandle, ByteSequenceTensorBuffer rawBuffer, Shape shape) {
+      this(tensorHandle, rawBuffer, shape, UTF_8_LAYOUT);
+    }
+
+    private TStringImpl(
+        TensorHandle tensorHandle,
+        ByteSequenceTensorBuffer rawBuffer,
+        Shape shape,
+        DataLayout<DataBuffer<byte[]>, String> layout
+    ) {
+      super(layout.applyTo(rawBuffer), shape);
+      this.rawBuffer = rawBuffer;
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+    private final ByteSequenceTensorBuffer rawBuffer;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TUint8Factory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TUint8Factory.java
new file mode 100644
index 00000000000..13d3e5bc663
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/internal/types/TUint8Factory.java
@@ -0,0 +1,41 @@
+package org.tensorflow.internal.types;
+
+import org.tensorflow.AbstractTypeFactory;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.internal.buffer.TensorBuffers;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.ByteDataBuffer;
+import org.tensorflow.ndarray.impl.dense.ByteDenseNdArray;
+import org.tensorflow.types.TUint8;
+
+/**
+ * Factory of {@link TUint8} tensor instances
+ */
+public class TUint8Factory extends AbstractTypeFactory<TUint8> {
+
+  @Override
+  public TUint8 createDense(TensorHandle tensorHandle, Shape shape) {
+    ByteDataBuffer buffer = TensorBuffers.toBytes(getNative(tensorHandle));
+    return new TUint8Impl(tensorHandle, buffer, shape);
+  }
+
+  private static final class TUint8Impl extends ByteDenseNdArray implements TUint8 {
+
+    @Override
+    public Class<TUint8> type() {
+      return TUint8.class;
+    }
+
+    @Override
+    public TensorHandle handle() {
+      return tensorHandle;
+    }
+
+    TUint8Impl(TensorHandle tensorHandle, ByteDataBuffer buffer, Shape shape) {
+      super(buffer, shape);
+      this.tensorHandle = tensorHandle;
+    }
+
+    private final TensorHandle tensorHandle;
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Constant.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Constant.java
index 6c214cc6819..3c2d34d3980 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Constant.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Constant.java
@@ -16,23 +16,10 @@
 package org.tensorflow.op.core;
 
 import java.nio.charset.Charset;
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.Output;
-import org.tensorflow.Tensor;
-import org.tensorflow.op.RawOp;
-import org.tensorflow.op.Scope;
-import org.tensorflow.op.annotation.Endpoint;
-import org.tensorflow.op.annotation.Operator;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.BooleanDataBuffer;
-import org.tensorflow.ndarray.buffer.ByteDataBuffer;
-import org.tensorflow.ndarray.buffer.DataBuffer;
-import org.tensorflow.ndarray.buffer.DoubleDataBuffer;
-import org.tensorflow.ndarray.buffer.FloatDataBuffer;
-import org.tensorflow.ndarray.buffer.IntDataBuffer;
-import org.tensorflow.ndarray.buffer.LongDataBuffer;
+import org.tensorflow.Tensors;
 import org.tensorflow.ndarray.BooleanNdArray;
 import org.tensorflow.ndarray.ByteNdArray;
 import org.tensorflow.ndarray.DoubleNdArray;
@@ -41,7 +28,19 @@
 import org.tensorflow.ndarray.LongNdArray;
 import org.tensorflow.ndarray.NdArray;
 import org.tensorflow.ndarray.NdArrays;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
+import org.tensorflow.ndarray.buffer.BooleanDataBuffer;
+import org.tensorflow.ndarray.buffer.ByteDataBuffer;
+import org.tensorflow.ndarray.buffer.DataBuffer;
+import org.tensorflow.ndarray.buffer.DoubleDataBuffer;
+import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.ndarray.buffer.IntDataBuffer;
+import org.tensorflow.ndarray.buffer.LongDataBuffer;
+import org.tensorflow.op.RawOp;
+import org.tensorflow.op.Scope;
+import org.tensorflow.op.annotation.Endpoint;
+import org.tensorflow.op.annotation.Operator;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TFloat64;
@@ -77,7 +76,7 @@ public final class Constant<T extends TType> extends RawOp implements Operand<T>
    */
   @Endpoint
   public static Constant<TInt32> scalarOf(Scope scope, int data) {
-    try (Tensor<TInt32> value = TInt32.scalarOf(data)) {
+    try (TInt32 value = TInt32.scalarOf(data)) {
       return create(scope, value);
     }
   }
@@ -92,7 +91,7 @@ public static Constant<TInt32> scalarOf(Scope scope, int data) {
    */
   @Endpoint
   public static Constant<TInt32> vectorOf(Scope scope, int[] data) {
-    try (Tensor<TInt32> value = TInt32.vectorOf(data)) {
+    try (TInt32 value = TInt32.vectorOf(data)) {
       return create(scope, value);
     }
   }
@@ -122,7 +121,7 @@ public static Constant<TInt32> arrayOf(Scope scope, int... data) {
    */
   @Endpoint
   public static Constant<TInt32> tensorOf(Scope scope, int[][] data) {
-    try (Tensor<TInt32> value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt32 value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -138,7 +137,7 @@ public static Constant<TInt32> tensorOf(Scope scope, int[][] data) {
    */
   @Endpoint
   public static Constant<TInt32> tensorOf(Scope scope, int[][][] data) {
-    try (Tensor<TInt32> value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt32 value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -154,7 +153,7 @@ public static Constant<TInt32> tensorOf(Scope scope, int[][][] data) {
    */
   @Endpoint
   public static Constant<TInt32> tensorOf(Scope scope, int[][][][] data) {
-    try (Tensor<TInt32> value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt32 value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -170,7 +169,7 @@ public static Constant<TInt32> tensorOf(Scope scope, int[][][][] data) {
    */
   @Endpoint
   public static Constant<TInt32> tensorOf(Scope scope, int[][][][][] data) {
-    try (Tensor<TInt32> value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt32 value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -186,7 +185,7 @@ public static Constant<TInt32> tensorOf(Scope scope, int[][][][][] data) {
    */
   @Endpoint
   public static Constant<TInt32> tensorOf(Scope scope, int[][][][][][] data) {
-    try (Tensor<TInt32> value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt32 value = TInt32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -201,7 +200,7 @@ public static Constant<TInt32> tensorOf(Scope scope, int[][][][][][] data) {
    */
   @Endpoint
   public static Constant<TInt32> tensorOf(Scope scope, IntNdArray data) {
-    try (Tensor<TInt32> value = TInt32.tensorOf(data)) {
+    try (TInt32 value = TInt32.tensorOf(data)) {
       return create(scope, value);
     }
   }
@@ -217,7 +216,7 @@ public static Constant<TInt32> tensorOf(Scope scope, IntNdArray data) {
    */
   @Endpoint
   public static Constant<TInt32> tensorOf(Scope scope, Shape shape, IntDataBuffer data) {
-    try (Tensor<TInt32> value = TInt32.tensorOf(shape, data)) {
+    try (TInt32 value = TInt32.tensorOf(shape, data)) {
       return create(scope, value);
     }
   }
@@ -231,7 +230,7 @@ public static Constant<TInt32> tensorOf(Scope scope, Shape shape, IntDataBuffer
    */
   @Endpoint
   public static Constant<TFloat32> scalarOf(Scope scope, float data) {
-    try (Tensor<TFloat32> value = TFloat32.scalarOf(data)) {
+    try (TFloat32 value = TFloat32.scalarOf(data)) {
       return create(scope, value);
     }
   }
@@ -246,7 +245,7 @@ public static Constant<TFloat32> scalarOf(Scope scope, float data) {
    */
   @Endpoint
   public static Constant<TFloat32> vectorOf(Scope scope, float[] data) {
-    try (Tensor<TFloat32> value = TFloat32.vectorOf(data)) {
+    try (TFloat32 value = TFloat32.vectorOf(data)) {
       return create(scope, value);
     }
   }
@@ -276,7 +275,7 @@ public static Constant<TFloat32> arrayOf(Scope scope, float... data) {
    */
   @Endpoint
   public static Constant<TFloat32> tensorOf(Scope scope, float[][] data) {
-    try (Tensor<TFloat32> value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat32 value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -292,7 +291,7 @@ public static Constant<TFloat32> tensorOf(Scope scope, float[][] data) {
    */
   @Endpoint
   public static Constant<TFloat32> tensorOf(Scope scope, float[][][] data) {
-    try (Tensor<TFloat32> value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat32 value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -308,7 +307,7 @@ public static Constant<TFloat32> tensorOf(Scope scope, float[][][] data) {
    */
   @Endpoint
   public static Constant<TFloat32> tensorOf(Scope scope, float[][][][] data) {
-    try (Tensor<TFloat32> value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat32 value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -324,7 +323,7 @@ public static Constant<TFloat32> tensorOf(Scope scope, float[][][][] data) {
    */
   @Endpoint
   public static Constant<TFloat32> tensorOf(Scope scope, float[][][][][] data) {
-    try (Tensor<TFloat32> value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat32 value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -340,7 +339,7 @@ public static Constant<TFloat32> tensorOf(Scope scope, float[][][][][] data) {
    */
   @Endpoint
   public static Constant<TFloat32> tensorOf(Scope scope, float[][][][][][] data) {
-    try (Tensor<TFloat32> value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat32 value = TFloat32.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -355,7 +354,7 @@ public static Constant<TFloat32> tensorOf(Scope scope, float[][][][][][] data) {
    */
   @Endpoint
   public static Constant<TFloat32> tensorOf(Scope scope, FloatNdArray data) {
-    try (Tensor<TFloat32> value = TFloat32.tensorOf(data)) {
+    try (TFloat32 value = TFloat32.tensorOf(data)) {
       return create(scope, value);
     }
   }
@@ -371,7 +370,7 @@ public static Constant<TFloat32> tensorOf(Scope scope, FloatNdArray data) {
    */
   @Endpoint
   public static Constant<TFloat32> tensorOf(Scope scope, Shape shape, FloatDataBuffer data) {
-    try (Tensor<TFloat32> value = TFloat32.tensorOf(shape, data)) {
+    try (TFloat32 value = TFloat32.tensorOf(shape, data)) {
       return create(scope, value);
     }
   }
@@ -385,7 +384,7 @@ public static Constant<TFloat32> tensorOf(Scope scope, Shape shape, FloatDataBuf
    */
   @Endpoint
   public static Constant<TFloat64> scalarOf(Scope scope, double data) {
-    try (Tensor<TFloat64> value = TFloat64.scalarOf(data)) {
+    try (TFloat64 value = TFloat64.scalarOf(data)) {
       return create(scope, value);
     }
   }
@@ -400,7 +399,7 @@ public static Constant<TFloat64> scalarOf(Scope scope, double data) {
    */
   @Endpoint
   public static Constant<TFloat64> vectorOf(Scope scope, double[] data) {
-    try (Tensor<TFloat64> value = TFloat64.vectorOf(data)) {
+    try (TFloat64 value = TFloat64.vectorOf(data)) {
       return create(scope, value);
     }
   }
@@ -430,7 +429,7 @@ public static Constant<TFloat64> arrayOf(Scope scope, double... data) {
    */
   @Endpoint
   public static Constant<TFloat64> tensorOf(Scope scope, double[][] data) {
-    try (Tensor<TFloat64> value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat64 value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -446,7 +445,7 @@ public static Constant<TFloat64> tensorOf(Scope scope, double[][] data) {
    */
   @Endpoint
   public static Constant<TFloat64> tensorOf(Scope scope, double[][][] data) {
-    try (Tensor<TFloat64> value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat64 value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -462,7 +461,7 @@ public static Constant<TFloat64> tensorOf(Scope scope, double[][][] data) {
    */
   @Endpoint
   public static Constant<TFloat64> tensorOf(Scope scope, double[][][][] data) {
-    try (Tensor<TFloat64> value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat64 value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -478,7 +477,7 @@ public static Constant<TFloat64> tensorOf(Scope scope, double[][][][] data) {
    */
   @Endpoint
   public static Constant<TFloat64> tensorOf(Scope scope, double[][][][][] data) {
-    try (Tensor<TFloat64> value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat64 value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -494,7 +493,7 @@ public static Constant<TFloat64> tensorOf(Scope scope, double[][][][][] data) {
    */
   @Endpoint
   public static Constant<TFloat64> tensorOf(Scope scope, double[][][][][][] data) {
-    try (Tensor<TFloat64> value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
+    try (TFloat64 value = TFloat64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(
         data, t))) {
       return create(scope, value);
     }
@@ -509,7 +508,7 @@ public static Constant<TFloat64> tensorOf(Scope scope, double[][][][][][] data)
    */
   @Endpoint
   public static Constant<TFloat64> tensorOf(Scope scope, DoubleNdArray data) {
-    try (Tensor<TFloat64> value = TFloat64.tensorOf(data)) {
+    try (TFloat64 value = TFloat64.tensorOf(data)) {
       return create(scope, value);
     }
   }
@@ -525,7 +524,7 @@ public static Constant<TFloat64> tensorOf(Scope scope, DoubleNdArray data) {
    */
   @Endpoint
   public static Constant<TFloat64> tensorOf(Scope scope, Shape shape, DoubleDataBuffer data) {
-    try (Tensor<TFloat64> value = TFloat64.tensorOf(shape, data)) {
+    try (TFloat64 value = TFloat64.tensorOf(shape, data)) {
       return create(scope, value);
     }
   }
@@ -539,7 +538,7 @@ public static Constant<TFloat64> tensorOf(Scope scope, Shape shape, DoubleDataBu
    */
   @Endpoint
   public static Constant<TInt64> scalarOf(Scope scope, long data) {
-    try (Tensor<TInt64> value = TInt64.scalarOf(data)) {
+    try (TInt64 value = TInt64.scalarOf(data)) {
       return create(scope, value);
     }
   }
@@ -554,7 +553,7 @@ public static Constant<TInt64> scalarOf(Scope scope, long data) {
    */
   @Endpoint
   public static Constant<TInt64> vectorOf(Scope scope, long[] data) {
-    try (Tensor<TInt64> value = TInt64.vectorOf(data)) {
+    try (TInt64 value = TInt64.vectorOf(data)) {
       return create(scope, value);
     }
   }
@@ -569,7 +568,7 @@ public static Constant<TInt64> vectorOf(Scope scope, long[] data) {
    */
   @Endpoint
   public static Constant<TInt64> tensorOf(Scope scope, long[][] data) {
-    try (Tensor<TInt64> value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt64 value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -600,7 +599,7 @@ public static Constant<TInt64> arrayOf(Scope scope, long... data) {
    */
   @Endpoint
   public static Constant<TInt64> tensorOf(Scope scope, long[][][] data) {
-    try (Tensor<TInt64> value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt64 value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -616,7 +615,7 @@ public static Constant<TInt64> tensorOf(Scope scope, long[][][] data) {
    */
   @Endpoint
   public static Constant<TInt64> tensorOf(Scope scope, long[][][][] data) {
-    try (Tensor<TInt64> value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt64 value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -632,7 +631,7 @@ public static Constant<TInt64> tensorOf(Scope scope, long[][][][] data) {
    */
   @Endpoint
   public static Constant<TInt64> tensorOf(Scope scope, long[][][][][] data) {
-    try (Tensor<TInt64> value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt64 value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -648,7 +647,7 @@ public static Constant<TInt64> tensorOf(Scope scope, long[][][][][] data) {
    */
   @Endpoint
   public static Constant<TInt64> tensorOf(Scope scope, long[][][][][][] data) {
-    try (Tensor<TInt64> value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TInt64 value = TInt64.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -663,7 +662,7 @@ public static Constant<TInt64> tensorOf(Scope scope, long[][][][][][] data) {
    */
   @Endpoint
   public static Constant<TInt64> tensorOf(Scope scope, LongNdArray data) {
-    try (Tensor<TInt64> value = TInt64.tensorOf(data)) {
+    try (TInt64 value = TInt64.tensorOf(data)) {
       return create(scope, value);
     }
   }
@@ -679,7 +678,7 @@ public static Constant<TInt64> tensorOf(Scope scope, LongNdArray data) {
    */
   @Endpoint
   public static Constant<TInt64> tensorOf(Scope scope, Shape shape, LongDataBuffer data) {
-    try (Tensor<TInt64> value = TInt64.tensorOf(shape, data)) {
+    try (TInt64 value = TInt64.tensorOf(shape, data)) {
       return create(scope, value);
     }
   }
@@ -693,7 +692,7 @@ public static Constant<TInt64> tensorOf(Scope scope, Shape shape, LongDataBuffer
    */
   @Endpoint
   public static Constant<TBool> scalarOf(Scope scope, boolean data) {
-    try (Tensor<TBool> value = TBool.scalarOf(data)) {
+    try (TBool value = TBool.scalarOf(data)) {
       return create(scope, value);
     }
   }
@@ -708,7 +707,7 @@ public static Constant<TBool> scalarOf(Scope scope, boolean data) {
    */
   @Endpoint
   public static Constant<TBool> vectorOf(Scope scope, boolean[] data) {
-    try (Tensor<TBool> value = TBool.vectorOf(data)) {
+    try (TBool value = TBool.vectorOf(data)) {
       return create(scope, value);
     }
   }
@@ -738,7 +737,7 @@ public static Constant<TBool> arrayOf(Scope scope, boolean... data) {
    */
   @Endpoint
   public static Constant<TBool> tensorOf(Scope scope, boolean[][] data) {
-    try (Tensor<TBool> value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TBool value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -754,7 +753,7 @@ public static Constant<TBool> tensorOf(Scope scope, boolean[][] data) {
    */
   @Endpoint
   public static Constant<TBool> tensorOf(Scope scope, boolean[][][] data) {
-    try (Tensor<TBool> value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TBool value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -770,7 +769,7 @@ public static Constant<TBool> tensorOf(Scope scope, boolean[][][] data) {
    */
   @Endpoint
   public static Constant<TBool> tensorOf(Scope scope, boolean[][][][] data) {
-    try (Tensor<TBool> value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TBool value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -786,7 +785,7 @@ public static Constant<TBool> tensorOf(Scope scope, boolean[][][][] data) {
    */
   @Endpoint
   public static Constant<TBool> tensorOf(Scope scope, boolean[][][][][] data) {
-    try (Tensor<TBool> value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TBool value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -802,7 +801,7 @@ public static Constant<TBool> tensorOf(Scope scope, boolean[][][][][] data) {
    */
   @Endpoint
   public static Constant<TBool> tensorOf(Scope scope, boolean[][][][][][] data) {
-    try (Tensor<TBool> value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
+    try (TBool value = TBool.tensorOf(StdArrays.shapeOf(data), t -> StdArrays.copyTo(data,
         t))) {
       return create(scope, value);
     }
@@ -817,7 +816,7 @@ public static Constant<TBool> tensorOf(Scope scope, boolean[][][][][][] data) {
    */
   @Endpoint
   public static Constant<TBool> tensorOf(Scope scope, BooleanNdArray data) {
-    try (Tensor<TBool> value = TBool.tensorOf(data)) {
+    try (TBool value = TBool.tensorOf(data)) {
       return create(scope, value);
     }
   }
@@ -833,7 +832,7 @@ public static Constant<TBool> tensorOf(Scope scope, BooleanNdArray data) {
    */
   @Endpoint
   public static Constant<TBool> tensorOf(Scope scope, Shape shape, BooleanDataBuffer data) {
-    try (Tensor<TBool> value = TBool.tensorOf(shape, data)) {
+    try (TBool value = TBool.tensorOf(shape, data)) {
       return create(scope, value);
     }
   }
@@ -847,7 +846,7 @@ public static Constant<TBool> tensorOf(Scope scope, Shape shape, BooleanDataBuff
    */
   @Endpoint
   public static Constant<TUint8> scalarOf(Scope scope, byte data) {
-    try (Tensor<TUint8> value = TUint8.scalarOf(data)) {
+    try (TUint8 value = TUint8.scalarOf(data)) {
       return create(scope, value);
     }
   }
@@ -862,7 +861,7 @@ public static Constant<TUint8> scalarOf(Scope scope, byte data) {
    */
   @Endpoint
   public static Constant<TUint8> vectorOf(Scope scope, byte[] data) {
-    try (Tensor<TUint8> value = TUint8.vectorOf(data)) {
+    try (TUint8 value = TUint8.vectorOf(data)) {
       return create(scope, value);
     }
   }
@@ -892,7 +891,7 @@ public static Constant<TUint8> arrayOf(Scope scope, byte... data) {
    */
   @Endpoint
   public static Constant<TUint8> tensorOf(Scope scope, byte[][] data) {
-    try (Tensor<TUint8> value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
+    try (TUint8 value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
         d))) {
       return create(scope, value);
     }
@@ -908,7 +907,7 @@ public static Constant<TUint8> tensorOf(Scope scope, byte[][] data) {
    */
   @Endpoint
   public static Constant<TUint8> tensorOf(Scope scope, byte[][][] data) {
-    try (Tensor<TUint8> value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
+    try (TUint8 value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
         d))) {
       return create(scope, value);
     }
@@ -924,7 +923,7 @@ public static Constant<TUint8> tensorOf(Scope scope, byte[][][] data) {
    */
   @Endpoint
   public static Constant<TUint8> tensorOf(Scope scope, byte[][][][] data) {
-    try (Tensor<TUint8> value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
+    try (TUint8 value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
         d))) {
       return create(scope, value);
     }
@@ -940,7 +939,7 @@ public static Constant<TUint8> tensorOf(Scope scope, byte[][][][] data) {
    */
   @Endpoint
   public static Constant<TUint8> tensorOf(Scope scope, byte[][][][][] data) {
-    try (Tensor<TUint8> value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
+    try (TUint8 value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
         d))) {
       return create(scope, value);
     }
@@ -956,7 +955,7 @@ public static Constant<TUint8> tensorOf(Scope scope, byte[][][][][] data) {
    */
   @Endpoint
   public static Constant<TUint8> tensorOf(Scope scope, byte[][][][][][] data) {
-    try (Tensor<TUint8> value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
+    try (TUint8 value = TUint8.tensorOf(StdArrays.shapeOf(data), d -> StdArrays.copyTo(data,
         d))) {
       return create(scope, value);
     }
@@ -971,7 +970,7 @@ public static Constant<TUint8> tensorOf(Scope scope, byte[][][][][][] data) {
    */
   @Endpoint
   public static Constant<TUint8> tensorOf(Scope scope, ByteNdArray data) {
-    try (Tensor<TUint8> value = TUint8.tensorOf(data)) {
+    try (TUint8 value = TUint8.tensorOf(data)) {
       return create(scope, value);
     }
   }
@@ -987,7 +986,7 @@ public static Constant<TUint8> tensorOf(Scope scope, ByteNdArray data) {
    */
   @Endpoint
   public static Constant<TUint8> tensorOf(Scope scope, Shape shape, ByteDataBuffer data) {
-    try (Tensor<TUint8> value = TUint8.tensorOf(shape, data)) {
+    try (TUint8 value = TUint8.tensorOf(shape, data)) {
       return create(scope, value);
     }
   }
@@ -996,7 +995,7 @@ public static Constant<TUint8> tensorOf(Scope scope, Shape shape, ByteDataBuffer
    * Create a constant with data from the given buffer.
    *
    * @param scope is a scope used to add the underlying operation.
-   * @param type the tensor datatype.
+   * @param type the tensor type.
    * @param shape the tensor shape.
    * @param data a buffer containing the tensor data.
    * @return a constant of type `type`
@@ -1004,9 +1003,9 @@ public static Constant<TUint8> tensorOf(Scope scope, Shape shape, ByteDataBuffer
    *     buffer
    */
   @Endpoint
-  public static <T extends TType> Constant<T> tensorOf(Scope scope, DataType<T> type, Shape shape,
+  public static <T extends TType> Constant<T> tensorOf(Scope scope, Class<T> type, Shape shape,
       ByteDataBuffer data) {
-    try (Tensor<T> value = Tensor.of(type, shape, data)) {
+    try (T value = Tensors.of(type, shape, data)) {
       return create(scope, value);
     }
   }
@@ -1020,7 +1019,7 @@ public static <T extends TType> Constant<T> tensorOf(Scope scope, DataType<T> ty
    */
   @Endpoint
   public static Constant<TString> scalarOf(Scope scope, String data) {
-    try (Tensor<TString> value = TString.scalarOf(data)) {
+    try (TString value = TString.scalarOf(data)) {
       return create(scope, value);
     }
   }
@@ -1035,7 +1034,7 @@ public static Constant<TString> scalarOf(Scope scope, String data) {
    */
   @Endpoint
   public static Constant<TString> scalarOf(Scope scope, Charset charset, String data) {
-    try (Tensor<TString> value = TString.tensorOf(charset, NdArrays.scalarOfObject(data))) {
+    try (TString value = TString.tensorOf(charset, NdArrays.scalarOfObject(data))) {
       return create(scope, value);
     }
   }
@@ -1049,7 +1048,7 @@ public static Constant<TString> scalarOf(Scope scope, Charset charset, String da
    */
   public static Constant<TString> vectorOf(Scope scope, String[] data) {
     NdArray<String> src = NdArrays.vectorOfObjects(data);
-    try (Tensor<TString> value = TString.tensorOf(src)) {
+    try (TString value = TString.tensorOf(src)) {
       return create(scope, value);
     }
   }
@@ -1065,7 +1064,7 @@ public static Constant<TString> vectorOf(Scope scope, String[] data) {
    */
   @Endpoint
   public static Constant<TString> vectorOf(Scope scope, Charset charset, String[] data) {
-    try (Tensor<TString> value = TString.tensorOf(charset, NdArrays.vectorOfObjects(data))) {
+    try (TString value = TString.tensorOf(charset, NdArrays.vectorOfObjects(data))) {
       return Constant.create(scope, value);
     }
   }
@@ -1112,7 +1111,7 @@ public static Constant<TString> arrayOf(Scope scope, Charset charset, String...
   public static Constant<TString> tensorOf(Scope scope, String[][] data) {
     NdArray<String> src = NdArrays.ofObjects(String.class, StdArrays.shapeOf(data));
     StdArrays.copyTo(data, src);
-    try (Tensor<TString> value = TString.tensorOf(src)) {
+    try (TString value = TString.tensorOf(src)) {
       return create(scope, value);
     }
   }
@@ -1127,7 +1126,7 @@ public static Constant<TString> tensorOf(Scope scope, String[][] data) {
   public static Constant<TString> tensorOf(Scope scope, String[][][] data) {
     NdArray<String> src = NdArrays.ofObjects(String.class, StdArrays.shapeOf(data));
     StdArrays.copyTo(data, src);
-    try (Tensor<TString> value = TString.tensorOf(src)) {
+    try (TString value = TString.tensorOf(src)) {
       return create(scope, value);
     }
   }
@@ -1142,7 +1141,7 @@ public static Constant<TString> tensorOf(Scope scope, String[][][] data) {
   public static Constant<TString> tensorOf(Scope scope, String[][][][] data) {
     NdArray<String> src = NdArrays.ofObjects(String.class, StdArrays.shapeOf(data));
     StdArrays.copyTo(data, src);
-    try (Tensor<TString> value = TString.tensorOf(src)) {
+    try (TString value = TString.tensorOf(src)) {
       return create(scope, value);
     }
   }
@@ -1157,7 +1156,7 @@ public static Constant<TString> tensorOf(Scope scope, String[][][][] data) {
   public static Constant<TString> tensorOf(Scope scope, String[][][][][] data) {
     NdArray<String> src = NdArrays.ofObjects(String.class, StdArrays.shapeOf(data));
     StdArrays.copyTo(data, src);
-    try (Tensor<TString> value = TString.tensorOf(src)) {
+    try (TString value = TString.tensorOf(src)) {
       return create(scope, value);
     }
   }
@@ -1172,7 +1171,7 @@ public static Constant<TString> tensorOf(Scope scope, String[][][][][] data) {
   public static Constant<TString> tensorOf(Scope scope, String[][][][][][] data) {
     NdArray<String> src = NdArrays.ofObjects(String.class, StdArrays.shapeOf(data));
     StdArrays.copyTo(data, src);
-    try (Tensor<TString> value = TString.tensorOf(src)) {
+    try (TString value = TString.tensorOf(src)) {
       return create(scope, value);
     }
   }
@@ -1187,7 +1186,7 @@ public static Constant<TString> tensorOf(Scope scope, String[][][][][][] data) {
    */
   @Endpoint
   public static Constant<TString> tensorOf(Scope scope, NdArray<String> data) {
-    try (Tensor<TString> value = TString.tensorOf(data)) {
+    try (TString value = TString.tensorOf(data)) {
       return create(scope, value);
     }
   }
@@ -1203,7 +1202,7 @@ public static Constant<TString> tensorOf(Scope scope, NdArray<String> data) {
    */
   @Endpoint
   public static Constant<TString> tensorOf(Scope scope, Charset charset, NdArray<String> data) {
-    try (Tensor<TString> value = TString.tensorOf(charset, data)) {
+    try (TString value = TString.tensorOf(charset, data)) {
       return create(scope, value);
     }
   }
@@ -1220,7 +1219,7 @@ public static Constant<TString> tensorOf(Scope scope, Charset charset, NdArray<S
    */
   @Endpoint
   public static Constant<TString> tensorOf(Scope scope, Shape shape, DataBuffer<String> data) {
-    try (Tensor<TString> value = TString.tensorOf(shape, data)) {
+    try (TString value = TString.tensorOf(shape, data)) {
       return create(scope, value);
     }
   }
@@ -1238,7 +1237,7 @@ public static Constant<TString> tensorOf(Scope scope, Shape shape, DataBuffer<St
   @Endpoint
   public static Constant<TString> tensorOf(Scope scope, Charset charset, Shape shape,
       DataBuffer<String> data) {
-    try (Tensor<TString> value = TString.tensorOf(charset, shape, data)) {
+    try (TString value = TString.tensorOf(charset, shape, data)) {
       return create(scope, value);
     }
   }
@@ -1264,13 +1263,13 @@ public static Constant<TInt64> tensorOf(Scope scope, Shape shape) {
    * @return a constant of the same data type as `tensor`
    */
   @Endpoint
-  public static <T extends TType> Constant<T> create(Scope scope, Tensor<T> tensor) {
+  public static <T extends TType> Constant<T> create(Scope scope, T tensor) {
     return new Constant<>(
         scope
             .env()
             .opBuilder("Const", scope.makeOpName("Const"))
             .setAttr("value", tensor)
-            .setAttr("dtype", tensor.dataType())
+            .setAttr("dtype", tensor.type())
             .build());
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Gradients.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Gradients.java
index 2827276c32c..45ac0d1180e 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Gradients.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Gradients.java
@@ -22,7 +22,6 @@
 import org.tensorflow.Graph;
 import org.tensorflow.Operand;
 import org.tensorflow.Output;
-import org.tensorflow.op.Op;
 import org.tensorflow.op.Operands;
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
@@ -101,9 +100,11 @@ public static Gradients create(
         }
       }
     }
-    Output<?>[] dy =
-        graph.addGradients(
-            scope.makeOpName("Gradients"), Operands.asOutputs(y), Operands.asOutputs(x), dx);
+    Output<?>[] dy = graph.addGradients(
+        scope.makeOpName("Gradients"),
+        Operands.asOutputs(y),
+        Operands.asOutputs(x), dx
+    );
     return new Gradients(Arrays.asList(dy));
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Helpers.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Helpers.java
index f9ce837fe60..6a3c3521997 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Helpers.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Helpers.java
@@ -47,7 +47,7 @@ private Helpers() {}
     @Endpoint(name = "variable")
     public static <T extends TType> Variable<T> createVariableWithInit(Scope scope, Operand<T> init, Variable.Options... options) {
         Output<T> initOutput = init.asOutput();
-        Variable<T> newVar = Variable.create(scope,initOutput.shape(), initOutput.dataType(), options);
+        Variable<T> newVar = Variable.create(scope,initOutput.shape(), initOutput.type(), options);
         Assign<T> assignOp = Assign.create(scope, newVar, init);
         Init.add(scope, assignOp);
         return newVar;
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Shapes.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Shapes.java
index 613cb729341..a85fdbc1ab2 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Shapes.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Shapes.java
@@ -15,16 +15,12 @@
 package org.tensorflow.op.core;
 
 import java.util.Arrays;
-
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
-
-import org.tensorflow.op.math.FloorMod;
-
 import org.tensorflow.op.dtypes.Cast;
+import org.tensorflow.op.math.FloorMod;
 import org.tensorflow.op.math.NotEqual;
 import org.tensorflow.op.math.Sub;
 import org.tensorflow.types.TBool;
@@ -51,8 +47,8 @@
  * Operand<TInt32> numPred = tf.shape.size(predShape, tf.constant(0));
  * Operand<TFloat32> predFlat = tf.shape.flatten(yPred);
  *
- * Shape<TInt64> predShape64 = tf.shape(yPred, TInt64.DTYPE);
- * Operand<TInt64> predSqueezed = tf.shape.squeeze(predShape64, TInt64.DTYPE);
+ * Shape<TInt64> predShape64 = tf.shape(yPred, TInt64.class);
+ * Operand<TInt64> predSqueezed = tf.shape.squeeze(predShape64, TInt64.class);
  * }</pre>
  */
 @Operator(group = "shape")
@@ -68,7 +64,7 @@ public abstract class Shapes {
    */
   @Endpoint(name = "flatten")
   public static <T extends TType> Operand<T> flatten(Scope scope, Operand<T> operand) {
-    return flatten(scope, operand, TInt32.DTYPE);
+    return flatten(scope, operand, TInt32.class);
   }
 
   /**
@@ -83,7 +79,7 @@ public static <T extends TType> Operand<T> flatten(Scope scope, Operand<T> opera
    */
   @Endpoint(name = "flatten")
   public static <T extends TType, U extends TNumber> Operand<T> flatten(
-      Scope scope, Operand<T> operand, DataType<U> dType) {
+      Scope scope, Operand<T> operand, Class<U> dType) {
     Operand<U> flatShape = flatten(scope, Shape.create(scope, operand, dType), dType);
     return Reshape.create(scope, operand, flatShape);
   }
@@ -97,7 +93,7 @@ public static <T extends TType, U extends TNumber> Operand<T> flatten(
    */
   @Endpoint(name = "flatten")
   public static Operand<TInt32> flatten(Scope scope, Shape<TInt32> shape) {
-    return flatten(scope, shape, TInt32.DTYPE);
+    return flatten(scope, shape, TInt32.class);
   }
 
   /**
@@ -111,11 +107,11 @@ public static Operand<TInt32> flatten(Scope scope, Shape<TInt32> shape) {
    */
   @Endpoint(name = "flatten")
   public static <U extends TNumber> Operand<U> flatten(
-      Scope scope, Shape<U> shape, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Class<U> dType) {
     return ExpandDims.create(
         scope,
         size(scope, shape, dType),
-        Cast.create(scope, Constant.scalarOf(scope, -1), TInt32.DTYPE));
+        Cast.create(scope, Constant.scalarOf(scope, -1), TInt32.class));
   }
 
   /**
@@ -127,7 +123,7 @@ public static <U extends TNumber> Operand<U> flatten(
    */
   @Endpoint(name = "size")
   public static Operand<TInt32> size(Scope scope, Shape<TInt32> shape) {
-    return size(scope, shape, TInt32.DTYPE);
+    return size(scope, shape, TInt32.class);
   }
 
   /**
@@ -141,7 +137,7 @@ public static Operand<TInt32> size(Scope scope, Shape<TInt32> shape) {
    */
   @Endpoint(name = "size")
   public static <U extends TNumber> Operand<U> size(
-      Scope scope, Shape<U> shape, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Class<U> dType) {
     Slice<U> dims =
         Slice.create(
             scope,
@@ -164,7 +160,7 @@ public static <U extends TNumber> Operand<U> size(
    */
   @Endpoint(name = "size")
   public static Operand<TInt32> size(Scope scope, Shape<TInt32> shape, Operand<TInt32> dim) {
-    return size(scope, shape, dim, TInt32.DTYPE);
+    return size(scope, shape, dim, TInt32.class);
   }
 
   /**
@@ -179,7 +175,7 @@ public static Operand<TInt32> size(Scope scope, Shape<TInt32> shape, Operand<TIn
    */
   @Endpoint(name = "size")
   public static <U extends TNumber> Operand<U> size(
-      Scope scope, Shape<U> shape, Operand<U> dim, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Operand<U> dim, Class<U> dType) {
     return Slice.create(
         scope,
         shape,
@@ -201,7 +197,7 @@ public static <U extends TNumber> Operand<U> size(
   @Endpoint(name = "size")
   public static <T extends TType> Operand<TInt32> size(
       Scope scope, Operand<T> input, Operand<TInt32> dim) {
-    return size(scope, input, dim, TInt32.DTYPE);
+    return size(scope, input, dim, TInt32.class);
   }
 
   /**
@@ -216,7 +212,7 @@ public static <T extends TType> Operand<TInt32> size(
    */
   @Endpoint(name = "size")
   public static <T extends TType, U extends TNumber> Operand<U> size(
-      Scope scope, Operand<T> input, Operand<U> dim, DataType<U> dType) {
+      Scope scope, Operand<T> input, Operand<U> dim, Class<U> dType) {
     return size(scope, Shape.create(scope, input, dType), dim, dType);
   }
 
@@ -229,7 +225,7 @@ public static <T extends TType, U extends TNumber> Operand<U> size(
    */
   @Endpoint(name = "numDimensions")
   public static Operand<TInt32> numDimensions(Scope scope, Shape<TInt32> shape) {
-    return Size.create(scope, shape, TInt32.DTYPE);
+    return Size.create(scope, shape, TInt32.class);
   }
 
   /**
@@ -243,7 +239,7 @@ public static Operand<TInt32> numDimensions(Scope scope, Shape<TInt32> shape) {
    */
   @Endpoint(name = "numDimensions")
   public static <U extends TNumber> Operand<U> numDimensions(
-      Scope scope, Shape<U> shape, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Class<U> dType) {
     return Size.create(scope, shape, dType);
   }
 
@@ -259,7 +255,7 @@ public static <U extends TNumber> Operand<U> numDimensions(
   @Endpoint(name = "reduceDims")
   public static <T extends TType> Operand<T> reduceDims(
       Scope scope, Operand<T> operand, Operand<TInt32> axis) {
-    return reduceDims(scope, operand, axis, TInt32.DTYPE);
+    return reduceDims(scope, operand, axis, TInt32.class);
   }
 
   /**
@@ -275,7 +271,7 @@ public static <T extends TType> Operand<T> reduceDims(
    */
   @Endpoint(name = "reduceDims")
   public static <T extends TType, U extends TNumber> Operand<T> reduceDims(
-      Scope scope, Operand<T> operand, Operand<U> axis, DataType<U> dType) {
+      Scope scope, Operand<T> operand, Operand<U> axis, Class<U> dType) {
     Shape<U> newShape = Shape.create(scope, operand, dType);
     return Reshape.create(scope, operand, reduceDims(scope, newShape, axis, dType));
   }
@@ -290,7 +286,7 @@ public static <T extends TType, U extends TNumber> Operand<T> reduceDims(
    */
   @Endpoint(name = "reduceDims")
   public static Operand<TInt32> reduceDims(Scope scope, Shape<TInt32> shape, Operand<TInt32> axis) {
-    return reduceDims(scope, shape, axis, TInt32.DTYPE);
+    return reduceDims(scope, shape, axis, TInt32.class);
   }
 
   /**
@@ -305,7 +301,7 @@ public static Operand<TInt32> reduceDims(Scope scope, Shape<TInt32> shape, Opera
    */
   @Endpoint(name = "reduceDims")
   public static <U extends TNumber> Operand<U> reduceDims(
-      Scope scope, Shape<U> shape, Operand<U> axis, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Operand<U> axis, Class<U> dType) {
     Size<U> rank = Size.create(scope, shape, dType);
     axis = FloorMod.create(scope, axis, rank);
     Sub<U> remainder = Sub.create(scope, rank, axis);
@@ -343,7 +339,7 @@ public static <U extends TNumber> Operand<U> reduceDims(
    */
   @Endpoint(name = "squeeze")
   public static Operand<TInt32> squeeze(Scope scope, Shape<TInt32> shape) {
-    return squeeze(scope, shape, TInt32.DTYPE);
+    return squeeze(scope, shape, TInt32.class);
   }
 
   /**
@@ -357,7 +353,7 @@ public static Operand<TInt32> squeeze(Scope scope, Shape<TInt32> shape) {
    */
   @Endpoint(name = "squeeze")
   public static <U extends TNumber> Operand<U> squeeze(
-      Scope scope, Shape<U> shape, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Class<U> dType) {
     Operand<TBool> mask =
         NotEqual.create(scope, shape, Cast.create(scope, OnesLike.create(scope, shape), dType));
 
@@ -373,7 +369,7 @@ public static <U extends TNumber> Operand<U> squeeze(
    */
   @Endpoint(name = "head")
   public static Operand<TInt32> head(Scope scope, Shape<TInt32> shape) {
-    return head(scope, shape, TInt32.DTYPE);
+    return head(scope, shape, TInt32.class);
   }
 
   /**
@@ -387,7 +383,7 @@ public static Operand<TInt32> head(Scope scope, Shape<TInt32> shape) {
    */
   @Endpoint(name = "head")
   public static <U extends TNumber> Operand<U> head(
-      Scope scope, Shape<U> shape, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Class<U> dType) {
     return take(scope, shape, Cast.create(scope, Constant.scalarOf(scope, 1), dType), dType);
   }
 
@@ -403,7 +399,7 @@ public static <U extends TNumber> Operand<U> head(
    */
   @Endpoint(name = "take")
   public static Operand<TInt32> take(Scope scope, Shape<TInt32> shape, Operand<TInt32> n) {
-    return take(scope, shape, n, TInt32.DTYPE);
+    return take(scope, shape, n, TInt32.class);
   }
 
   /**
@@ -420,7 +416,7 @@ public static Operand<TInt32> take(Scope scope, Shape<TInt32> shape, Operand<TIn
    */
   @Endpoint(name = "take")
   public static <U extends TNumber> Operand<U> take(
-      Scope scope, Shape<U> shape, Operand<U> n, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Operand<U> n, Class<U> dType) {
     return Slice.create(
         scope,
         shape,
@@ -439,7 +435,7 @@ public static <U extends TNumber> Operand<U> take(
    */
   @Endpoint(name = "tail")
   public static Operand<TInt32> tail(Scope scope, Shape<TInt32> shape) {
-    return tail(scope, shape, TInt32.DTYPE);
+    return tail(scope, shape, TInt32.class);
   }
 
   /**
@@ -455,7 +451,7 @@ public static Operand<TInt32> tail(Scope scope, Shape<TInt32> shape) {
    */
   @Endpoint(name = "tail")
   public static <U extends TNumber> Operand<U> tail(
-      Scope scope, Shape<U> shape, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Class<U> dType) {
     return takeLast(scope, shape, Cast.create(scope, Constant.scalarOf(scope, 1), dType), dType);
   }
 
@@ -472,7 +468,7 @@ public static <U extends TNumber> Operand<U> tail(
   @Endpoint(name = "takeLast")
   public static <U extends TNumber> Operand<TInt32> takeLast(
       Scope scope, Shape<TInt32> shape, Operand<TInt32> n) {
-    return takeLast(scope, shape, n, TInt32.DTYPE);
+    return takeLast(scope, shape, n, TInt32.class);
   }
 
   /**
@@ -489,7 +485,7 @@ public static <U extends TNumber> Operand<TInt32> takeLast(
    */
   @Endpoint(name = "takeLast")
   public static <U extends TNumber> Operand<U> takeLast(
-      Scope scope, Shape<U> shape, Operand<U> n, DataType<U> dType) {
+      Scope scope, Shape<U> shape, Operand<U> n, Class<U> dType) {
 
     Size<U> rank = Size.create(scope, shape, dType);
     Sub<U> start = Sub.create(scope, rank, n);
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Zeros.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Zeros.java
index 4aad417b117..3de8b1d26e8 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Zeros.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/core/Zeros.java
@@ -14,7 +14,6 @@
 ==============================================================================*/
 package org.tensorflow.op.core;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Operation;
 import org.tensorflow.Output;
@@ -51,10 +50,10 @@ public final class Zeros<T extends TType> implements Op, Operand<T> {
    */
   @Endpoint
   @SuppressWarnings("unchecked")
-  public static <T extends TType, U extends TNumber> Zeros<T> create(Scope scope, Operand<U> dims, DataType<T> type) {
+  public static <T extends TType, U extends TNumber> Zeros<T> create(Scope scope, Operand<U> dims, Class<T> type) {
     Scope zerosScope = scope.withSubScope("Zeros");
     Operand<T> zero;
-    if (type == TString.DTYPE) {
+    if (type == TString.class) {
       zero = (Operand<T>)Constant.scalarOf(zerosScope.withName("Zero"), "");
     } else {
       zero = Cast.create(zerosScope.withName("Zero"), Constant.scalarOf(zerosScope, 0), type);
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SigmoidCrossEntropyWithLogits.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SigmoidCrossEntropyWithLogits.java
index 4f3e9569103..191e4b3c122 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SigmoidCrossEntropyWithLogits.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SigmoidCrossEntropyWithLogits.java
@@ -8,7 +8,13 @@
 import org.tensorflow.op.core.Select;
 import org.tensorflow.op.core.ZerosLike;
 import org.tensorflow.op.dtypes.Cast;
-import org.tensorflow.op.math.*;
+import org.tensorflow.op.math.Add;
+import org.tensorflow.op.math.Exp;
+import org.tensorflow.op.math.GreaterEqual;
+import org.tensorflow.op.math.Log1p;
+import org.tensorflow.op.math.Mul;
+import org.tensorflow.op.math.Neg;
+import org.tensorflow.op.math.Sub;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.family.TNumber;
 
@@ -72,7 +78,7 @@ public static <T extends TNumber> Operand<T> sigmoidCrossEntropyWithLogits(
     scope = scope.withSubScope("SigmoidCrossEntropyWithLogits");
 
     Operand<T> zeros =
-        Cast.create(scope, ZerosLike.create(scope, logits), logits.asOutput().dataType());
+        Cast.create(scope, ZerosLike.create(scope, logits), logits.asOutput().type());
     Operand<TBool> cond = GreaterEqual.create(scope, logits, zeros);
 
     Operand<T> reluLogits = Select.create(scope, cond, logits, zeros);
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SoftmaxCrossEntropyWithLogits.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SoftmaxCrossEntropyWithLogits.java
index 0c8bac697ed..d958c1272b1 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SoftmaxCrossEntropyWithLogits.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SoftmaxCrossEntropyWithLogits.java
@@ -1,12 +1,18 @@
 package org.tensorflow.op.nn;
 
-import org.tensorflow.DataType;
+import java.util.Arrays;
+import java.util.List;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.Scope;
 import org.tensorflow.op.annotation.Endpoint;
 import org.tensorflow.op.annotation.Operator;
-import org.tensorflow.op.core.*;
+import org.tensorflow.op.core.Concat;
+import org.tensorflow.op.core.Constant;
+import org.tensorflow.op.core.Range;
+import org.tensorflow.op.core.Rank;
+import org.tensorflow.op.core.Reshape;
+import org.tensorflow.op.core.Slice;
 import org.tensorflow.op.dtypes.Cast;
 import org.tensorflow.op.linalg.Transpose;
 import org.tensorflow.op.math.Sub;
@@ -15,10 +21,6 @@
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TNumber;
-import org.tensorflow.types.family.TType;
-
-import java.util.Arrays;
-import java.util.List;
 
 @Operator(group = "nn")
 public class SoftmaxCrossEntropyWithLogits {
@@ -78,24 +80,20 @@ public static <T extends TNumber, U extends TNumber> Operand<T> softmaxCrossEntr
       axis += logits.asOutput().shape().numDimensions();
     }
 
-
-    boolean convertToFloat32 =
-        logits.asOutput().dataType() == TFloat16.DTYPE
-            || logits.asOutput().dataType() == TBfloat16.DTYPE;
-    if (convertToFloat32) {
+    if (logits.asOutput().type() == TFloat16.class || logits.asOutput().type() == TBfloat16.class) {
       Operand<TFloat32> result =  softmaxCrossEntropyWithLogits(scope,
-              Cast.create(scope, labels, TFloat32.DTYPE),
-              Cast.create(scope, logits, TFloat32.DTYPE),
+              Cast.create(scope, labels, TFloat32.class),
+              Cast.create(scope, logits, TFloat32.class),
               axis);
-      return Cast.create(scope, result, logits.asOutput().dataType());
-    } else if(!logits.asOutput().dataType().equals(labels.asOutput().dataType())) {
+      return Cast.create(scope, result, logits.asOutput().type());
+    } else if(!logits.asOutput().type().equals(labels.asOutput().type())) {
       return softmaxCrossEntropyWithLogits(scope,
-              Cast.create(scope, labels, logits.asOutput().dataType()),
+              Cast.create(scope, labels, logits.asOutput().type()),
               logits,
               axis);
     }
 
-    Operand<TInt64> inputRank = Cast.create(scope, Rank.create(scope, logits), TInt64.DTYPE);
+    Operand<TInt64> inputRank = Cast.create(scope, Rank.create(scope, logits), TInt64.class);
     Shape shape = logits.asOutput().shape();
 
     // Move the dim to the end if dim is not the last dimension.
@@ -167,13 +165,13 @@ private static <T extends TNumber> Operand<T> flattenOuterDims(Scope scope, Oper
       }
     }
 
-    Operand<TInt64> rank = Cast.create(scope, Rank.create(scope, logits), TInt64.DTYPE);
+    Operand<TInt64> rank = Cast.create(scope, Rank.create(scope, logits), TInt64.class);
     Operand<TInt64> rankMinusOne = Sub.create(scope, rank, one);
 
     Operand<TInt64> lastDimSize =
         Slice.create(
             scope,
-            org.tensorflow.op.core.Shape.create(scope, logits, TInt64.DTYPE),
+            org.tensorflow.op.core.Shape.create(scope, logits, TInt64.class),
             rankMinusOne,
             one);
     Operand<TInt64> concat =
@@ -197,7 +195,7 @@ private static <T extends TNumber> Operand<T> flattenOuterDims(Scope scope, Oper
    */
   private static <T extends TNumber, U extends TNumber> Operand<T> moveDimToEnd(
       Scope scope, Operand<T> input, int dimIndex, Operand<U> rank) {
-    DataType<? extends TNumber> rankDType = rank.asOutput().dataType();
+    Class<U> rankDType = rank.asOutput().type();
     Operand one = Cast.create(scope, Constant.scalarOf(scope, 1), rankDType);
     List<Operand<U>> concatList =
         Arrays.asList(
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SparseSoftmaxCrossEntropyWithLogits.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SparseSoftmaxCrossEntropyWithLogits.java
index ebd6f74e7d8..79840827c13 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SparseSoftmaxCrossEntropyWithLogits.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/op/nn/SparseSoftmaxCrossEntropyWithLogits.java
@@ -1,5 +1,8 @@
 package org.tensorflow.op.nn;
 
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.List;
 import org.tensorflow.Operand;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.Op;
@@ -18,10 +21,6 @@
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TNumber;
 
-import java.util.ArrayList;
-import java.util.Collections;
-import java.util.List;
-
 @Operator(group = "nn")
 public class SparseSoftmaxCrossEntropyWithLogits {
 
@@ -74,11 +73,8 @@ public static <T extends TNumber, U extends TNumber> Operand sparseSoftmaxCrossE
     scope = scope.withSubScope("SparseSoftmaxCrossEntropyWithLogits");
     /** cannot use generics on preciseLogits as it may be recast later */
     Operand preciseLogits = logits;
-    boolean convertToFloat32 =
-        logits.asOutput().dataType() == TFloat16.DTYPE
-            || logits.asOutput().dataType() == TBfloat16.DTYPE;
-    if (convertToFloat32) {
-      preciseLogits = Cast.create(scope, logits, TFloat32.DTYPE);
+    if (logits.asOutput().type() == TFloat16.class || logits.asOutput().type() == TBfloat16.class) {
+      preciseLogits = Cast.create(scope, logits, TFloat32.class);
     }
     Shape labelsStaticShape = labels.asOutput().shape();
     org.tensorflow.op.core.Shape<TInt32> labelsShape =
@@ -115,8 +111,8 @@ public static <T extends TNumber, U extends TNumber> Operand sparseSoftmaxCrossE
           org.tensorflow.op.nn.raw.SparseSoftmaxCrossEntropyWithLogits.create(
               scope, preciseLogits, labels);
       Operand loss = smax.loss();
-      if (logits.asOutput().dataType() == TFloat16.DTYPE) {
-        loss = Cast.create(scope, loss, TFloat16.DTYPE);
+      if (logits.asOutput().type() == TFloat16.class) {
+        loss = Cast.create(scope, loss, TFloat16.class);
       }
       return loss;
     }
@@ -153,8 +149,8 @@ public static <T extends TNumber, U extends TNumber> Operand sparseSoftmaxCrossE
             scope, preciseLogits, labels);
     Operand cost = smax.loss();
     cost = Reshape.create(scope, cost, labelsShape);
-    if (logits.asOutput().dataType() == TFloat16.DTYPE) {
-      cost = Cast.create(scope, cost, TFloat16.DTYPE);
+    if (logits.asOutput().type() == TFloat16.class) {
+      cost = Cast.create(scope, cost, TFloat16.class);
     }
     return cost;
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBfloat16.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBfloat16.java
index 50f6ea49b06..6b56c310e47 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBfloat16.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBfloat16.java
@@ -18,18 +18,16 @@
 package org.tensorflow.types;
 
 import java.util.function.Consumer;
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
+import org.tensorflow.Tensors;
 import org.tensorflow.exceptions.TensorFlowException;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.FloatDataBuffer;
-import org.tensorflow.ndarray.buffer.layout.DataLayouts;
+import org.tensorflow.internal.types.TBfloat16Factory;
 import org.tensorflow.ndarray.FloatNdArray;
 import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
-import org.tensorflow.ndarray.impl.dense.FloatDenseNdArray;
+import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
 import org.tensorflow.types.family.TFloating;
 
 /**
@@ -48,12 +46,8 @@
  * <p>Note that some CPUs support the bfloat16 format natively, which can result in faster
  * computation compared to {@link TFloat16} when GPUs are not used.
  */
-public interface TBfloat16 extends FloatNdArray, TFloating {
-  /** readable-name for the data type */
-  static final String NAME = "BFLOAT16";
-
-  /** Type metadata */
-  DataType<TBfloat16> DTYPE = DataType.create(NAME, 14, 2, TBfloat16Impl::mapTensor);
+@TensorType(dataType = DataType.DT_BFLOAT16, byteSize = 2, factory = TBfloat16Factory.class)
+public interface TBfloat16 extends TFloating, FloatNdArray {
 
   /**
    * Allocates a new tensor for storing a single float value.
@@ -61,8 +55,8 @@ public interface TBfloat16 extends FloatNdArray, TFloating {
    * @param value float to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TBfloat16> scalarOf(float value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setFloat(value));
+  static TBfloat16 scalarOf(float value) {
+    return Tensors.of(TBfloat16.class, Shape.scalar(), t -> t.setFloat(value));
   }
 
   /**
@@ -71,11 +65,11 @@ static Tensor<TBfloat16> scalarOf(float value) {
    * @param values floats to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TBfloat16> vectorOf(float... values) {
+  static TBfloat16 vectorOf(float... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TBfloat16.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -86,8 +80,8 @@ static Tensor<TBfloat16> vectorOf(float... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TBfloat16> tensorOf(NdArray<Float> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TBfloat16 tensorOf(NdArray<Float> src) {
+    return Tensors.of(TBfloat16.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -96,8 +90,8 @@ static Tensor<TBfloat16> tensorOf(NdArray<Float> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TBfloat16> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TBfloat16 tensorOf(Shape shape) {
+    return Tensors.of(TBfloat16.class, shape);
   }
 
   /**
@@ -107,32 +101,20 @@ static Tensor<TBfloat16> tensorOf(Shape shape) {
    * @param data buffer of floats to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TBfloat16> tensorOf(Shape shape, FloatDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TBfloat16 tensorOf(Shape shape, FloatDataBuffer data) {
+    return Tensors.of(TBfloat16.class, shape, t -> t.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    * @throws TensorFlowException if the tensor cannot be allocated or initialized
    */
-  static Tensor<TBfloat16> tensorOf(Shape shape, Consumer<TBfloat16> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
+  static TBfloat16 tensorOf(Shape shape, Consumer<TBfloat16> tensorInit) {
+    return Tensors.of(TBfloat16.class, shape, tensorInit);
   }
 }
 
-/** Hidden implementation of a {@code TBfloat16} */
-class TBfloat16Impl extends FloatDenseNdArray implements TBfloat16 {
-
-  static TBfloat16 mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TBfloat16Impl(
-        DataLayouts.BFLOAT16.applyTo(TensorBuffers.toShorts(nativeTensor)), shape);
-  }
-
-  private TBfloat16Impl(FloatDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBool.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBool.java
index 3cc72101893..f5e5976bc73 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBool.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TBool.java
@@ -17,22 +17,20 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
+import java.util.function.Consumer;
+import org.tensorflow.Tensors;
 import org.tensorflow.exceptions.TensorFlowException;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
+import org.tensorflow.internal.types.TBoolFactory;
 import org.tensorflow.ndarray.BooleanNdArray;
 import org.tensorflow.ndarray.NdArray;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
 import org.tensorflow.ndarray.buffer.BooleanDataBuffer;
 import org.tensorflow.ndarray.buffer.layout.DataLayouts;
-import org.tensorflow.ndarray.impl.dense.BooleanDenseNdArray;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
 import org.tensorflow.types.family.TType;
 
-import java.util.function.Consumer;
-
 /**
  * Boolean tensor type.
  *
@@ -40,12 +38,8 @@
  * explicit mapping between Java boolean values and byte buffers using the {@link DataLayouts#BOOL
  * BOOL} layout, which may impact I/O performances.
  */
-public interface TBool extends BooleanNdArray, TType {
-  /** readable-name for the data type */
-  static final String NAME = "BOOL";
-
-  /** Type metadata */
-  DataType<TBool> DTYPE = DataType.create(NAME, 10, 1, TBoolImpl::mapTensor);
+@TensorType(dataType = DataType.DT_BOOL, byteSize = 1, factory = TBoolFactory.class)
+public interface TBool extends TType, BooleanNdArray {
 
   /**
    * Allocates a new tensor for storing a single boolean value.
@@ -53,8 +47,8 @@ public interface TBool extends BooleanNdArray, TType {
    * @param value boolean to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TBool> scalarOf(boolean value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setBoolean(value));
+  static TBool scalarOf(boolean value) {
+    return Tensors.of(TBool.class, Shape.scalar(), t -> t.setBoolean(value));
   }
 
   /**
@@ -63,11 +57,11 @@ static Tensor<TBool> scalarOf(boolean value) {
    * @param values booleans to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TBool> vectorOf(boolean... values) {
+  static TBool vectorOf(boolean... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TBool.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -78,8 +72,8 @@ static Tensor<TBool> vectorOf(boolean... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TBool> tensorOf(NdArray<Boolean> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TBool tensorOf(NdArray<Boolean> src) {
+    return Tensors.of(TBool.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -88,8 +82,8 @@ static Tensor<TBool> tensorOf(NdArray<Boolean> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TBool> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TBool tensorOf(Shape shape) {
+    return Tensors.of(TBool.class, shape);
   }
 
   /**
@@ -99,31 +93,20 @@ static Tensor<TBool> tensorOf(Shape shape) {
    * @param data buffer of booleans to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TBool> tensorOf(Shape shape, BooleanDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TBool tensorOf(Shape shape, BooleanDataBuffer data) {
+    return Tensors.of(TBool.class, shape, d -> d.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    * @throws TensorFlowException if the tensor cannot be allocated or initialized
    */
-  static Tensor<TBool> tensorOf(Shape shape, Consumer<TBool> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
+  static TBool tensorOf(Shape shape, Consumer<TBool> tensorInit) {
+    return Tensors.of(TBool.class, shape, tensorInit);
   }
-}
 
-/** Hidden implementation of a {@code TBool} */
-class TBoolImpl extends BooleanDenseNdArray implements TBool {
-
-  static TBool mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TBoolImpl(TensorBuffers.toBooleans(nativeTensor), shape);
-  }
-
-  private TBoolImpl(BooleanDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
-  }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat16.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat16.java
index 0cd441a1ff1..7add4b25346 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat16.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat16.java
@@ -18,18 +18,16 @@
 package org.tensorflow.types;
 
 import java.util.function.Consumer;
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
+import org.tensorflow.Tensors;
 import org.tensorflow.exceptions.TensorFlowException;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.FloatDataBuffer;
-import org.tensorflow.ndarray.buffer.layout.DataLayouts;
+import org.tensorflow.internal.types.TFloat16Factory;
 import org.tensorflow.ndarray.FloatNdArray;
 import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
-import org.tensorflow.ndarray.impl.dense.FloatDenseNdArray;
+import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
 import org.tensorflow.types.family.TFloating;
 
 /**
@@ -45,13 +43,8 @@
  * most CPUs do not support this format natively. For CPU computation on 16-bit floats, the {@link
  * TBfloat16} tensor type might be a better option.
  */
-public interface TFloat16 extends FloatNdArray, TFloating {
-
-  /** readable-name for the data type */
-  static final String NAME = "FLOAT16";
-
-  /** Type metadata */
-  DataType<TFloat16> DTYPE = DataType.create(NAME, 19, 2, TFloat16Impl::mapTensor);
+@TensorType(dataType = DataType.DT_HALF, byteSize = 2, factory = TFloat16Factory.class)
+public interface TFloat16 extends TFloating, FloatNdArray {
 
   /**
    * Allocates a new tensor for storing a single float value.
@@ -59,8 +52,8 @@ public interface TFloat16 extends FloatNdArray, TFloating {
    * @param value float to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat16> scalarOf(float value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setFloat(value));
+  static TFloat16 scalarOf(float value) {
+    return Tensors.of(TFloat16.class, Shape.scalar(), t -> t.setFloat(value));
   }
 
   /**
@@ -69,11 +62,11 @@ static Tensor<TFloat16> scalarOf(float value) {
    * @param values floats to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat16> vectorOf(float... values) {
+  static TFloat16 vectorOf(float... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TFloat16.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -84,8 +77,8 @@ static Tensor<TFloat16> vectorOf(float... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat16> tensorOf(NdArray<Float> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TFloat16 tensorOf(NdArray<Float> src) {
+    return Tensors.of(TFloat16.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -94,8 +87,8 @@ static Tensor<TFloat16> tensorOf(NdArray<Float> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TFloat16> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TFloat16 tensorOf(Shape shape) {
+    return Tensors.of(TFloat16.class, shape);
   }
 
   /**
@@ -105,32 +98,20 @@ static Tensor<TFloat16> tensorOf(Shape shape) {
    * @param data buffer of floats to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TFloat16> tensorOf(Shape shape, FloatDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TFloat16 tensorOf(Shape shape, FloatDataBuffer data) {
+    return Tensors.of(TFloat16.class, shape, t -> t.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    * @throws TensorFlowException if the tensor cannot be allocated or initialized
    */
-  static Tensor<TFloat16> tensorOf(Shape shape, Consumer<TFloat16> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
+  static TFloat16 tensorOf(Shape shape, Consumer<TFloat16> tensorInit) {
+    return Tensors.of(TFloat16.class, shape, tensorInit);
   }
-}
-
-/** Hidden implementation of a {@code TFloat16} */
-class TFloat16Impl extends FloatDenseNdArray implements TFloat16 {
 
-  static TFloat16 mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TFloat16Impl(
-        DataLayouts.FLOAT16.applyTo(TensorBuffers.toShorts(nativeTensor)), shape);
-  }
-
-  private TFloat16Impl(FloatDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
-  }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat32.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat32.java
index 571ec118ddc..32bc85ffa3e 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat32.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat32.java
@@ -18,27 +18,23 @@
 package org.tensorflow.types;
 
 import java.util.function.Consumer;
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
+import org.tensorflow.Tensors;
 import org.tensorflow.exceptions.TensorFlowException;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.internal.types.TFloat32Factory;
 import org.tensorflow.ndarray.FloatNdArray;
 import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
-import org.tensorflow.ndarray.impl.dense.FloatDenseNdArray;
+import org.tensorflow.ndarray.buffer.FloatDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
 import org.tensorflow.types.family.TFloating;
 
-/** IEEE-754 single-precision 32-bit float tensor type. */
-public interface TFloat32 extends FloatNdArray, TFloating {
-
-  /** readable-name for the data type */
-  static final String NAME = "FLOAT";
-
-  /** Type metadata */
-  DataType<TFloat32> DTYPE = DataType.create(NAME, 1, 4, TFloat32Impl::mapTensor);
+/**
+ * IEEE-754 single-precision 32-bit float tensor type.
+ */
+@TensorType(dataType = DataType.DT_FLOAT, byteSize = 4, factory = TFloat32Factory.class)
+public interface TFloat32 extends TFloating, FloatNdArray {
 
   /**
    * Allocates a new tensor for storing a single float value.
@@ -46,8 +42,8 @@ public interface TFloat32 extends FloatNdArray, TFloating {
    * @param value float to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat32> scalarOf(float value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setFloat(value));
+  static TFloat32 scalarOf(float value) {
+    return Tensors.of(TFloat32.class, Shape.scalar(), t -> t.setFloat(value));
   }
 
   /**
@@ -56,11 +52,11 @@ static Tensor<TFloat32> scalarOf(float value) {
    * @param values floats to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat32> vectorOf(float... values) {
+  static TFloat32 vectorOf(float... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TFloat32.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -71,8 +67,8 @@ static Tensor<TFloat32> vectorOf(float... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat32> tensorOf(NdArray<Float> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TFloat32 tensorOf(NdArray<Float> src) {
+    return Tensors.of(TFloat32.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -81,8 +77,8 @@ static Tensor<TFloat32> tensorOf(NdArray<Float> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TFloat32> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TFloat32 tensorOf(Shape shape) {
+    return Tensors.of(TFloat32.class, shape);
   }
 
   /**
@@ -92,31 +88,20 @@ static Tensor<TFloat32> tensorOf(Shape shape) {
    * @param data buffer of floats to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TFloat32> tensorOf(Shape shape, FloatDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TFloat32 tensorOf(Shape shape, FloatDataBuffer data) {
+    return Tensors.of(TFloat32.class, shape, t -> t.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    * @throws TensorFlowException if the tensor cannot be allocated or initialized
    */
-  static Tensor<TFloat32> tensorOf(Shape shape, Consumer<TFloat32> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
-  }
-}
-
-/** Hidden implementation of a {@code TFloat32} */
-class TFloat32Impl extends FloatDenseNdArray implements TFloat32 {
-
-  static TFloat32 mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TFloat32Impl(TensorBuffers.toFloats(nativeTensor), shape);
+  static TFloat32 tensorOf(Shape shape, Consumer<TFloat32> tensorInit) {
+    return Tensors.of(TFloat32.class, shape, tensorInit);
   }
 
-  private TFloat32Impl(FloatDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
-  }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat64.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat64.java
index 5d2744c4b3c..cc5799bf27f 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat64.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TFloat64.java
@@ -18,28 +18,23 @@
 package org.tensorflow.types;
 
 import java.util.function.Consumer;
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
+import org.tensorflow.Tensors;
 import org.tensorflow.exceptions.TensorFlowException;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.DoubleDataBuffer;
+import org.tensorflow.internal.types.TFloat64Factory;
 import org.tensorflow.ndarray.DoubleNdArray;
 import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
-import org.tensorflow.ndarray.impl.dense.DoubleDenseNdArray;
+import org.tensorflow.ndarray.buffer.DoubleDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
 import org.tensorflow.types.family.TFloating;
 
-
-/** IEEE-754 double-precision 64-bit float tensor type. */
-public interface TFloat64 extends DoubleNdArray, TFloating {
-
-  /** readable-name for the data type */
-  static final String NAME = "DOUBLE";
-
-  /** Type metadata */
-  DataType<TFloat64> DTYPE = DataType.create(NAME, 2, 8, TFloat64Impl::mapTensor);
+/**
+ * IEEE-754 double-precision 64-bit float tensor type.
+ */
+@TensorType(dataType = DataType.DT_DOUBLE, byteSize = 8, factory = TFloat64Factory.class)
+public interface TFloat64 extends TFloating, DoubleNdArray {
 
   /**
    * Allocates a new tensor for storing a single double value.
@@ -47,8 +42,8 @@ public interface TFloat64 extends DoubleNdArray, TFloating {
    * @param value double to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat64> scalarOf(double value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setDouble(value));
+  static TFloat64 scalarOf(double value) {
+    return Tensors.of(TFloat64.class, Shape.scalar(), t -> t.setDouble(value));
   }
 
   /**
@@ -57,11 +52,11 @@ static Tensor<TFloat64> scalarOf(double value) {
    * @param values doubles to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat64> vectorOf(double... values) {
+  static TFloat64 vectorOf(double... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TFloat64.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -72,8 +67,8 @@ static Tensor<TFloat64> vectorOf(double... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TFloat64> tensorOf(NdArray<Double> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TFloat64 tensorOf(NdArray<Double> src) {
+    return Tensors.of(TFloat64.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -82,8 +77,8 @@ static Tensor<TFloat64> tensorOf(NdArray<Double> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TFloat64> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TFloat64 tensorOf(Shape shape) {
+    return Tensors.of(TFloat64.class, shape);
   }
 
   /**
@@ -93,31 +88,19 @@ static Tensor<TFloat64> tensorOf(Shape shape) {
    * @param data buffer of doubles to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TFloat64> tensorOf(Shape shape, DoubleDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TFloat64 tensorOf(Shape shape, DoubleDataBuffer data) {
+    return Tensors.of(TFloat64.class, shape, t -> t.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    * @throws TensorFlowException if the tensor cannot be allocated or initialized
    */
-  static Tensor<TFloat64> tensorOf(Shape shape, Consumer<TFloat64> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
-  }
-}
-
-/** Hidden implementation of a {@code TFloat64} */
-class TFloat64Impl extends DoubleDenseNdArray implements TFloat64 {
-
-  static TFloat64 mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TFloat64Impl(TensorBuffers.toDoubles(nativeTensor), shape);
-  }
-
-  private TFloat64Impl(DoubleDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
+  static TFloat64 tensorOf(Shape shape, Consumer<TFloat64> tensorInit) {
+    return Tensors.of(TFloat64.class, shape, tensorInit);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt32.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt32.java
index 4a1139ddde2..ff2f7277ab8 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt32.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt32.java
@@ -18,26 +18,23 @@
 package org.tensorflow.types;
 
 import java.util.function.Consumer;
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.IntDataBuffer;
+import org.tensorflow.Tensors;
+import org.tensorflow.internal.types.TInt32Factory;
 import org.tensorflow.ndarray.IntNdArray;
 import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
-import org.tensorflow.ndarray.impl.dense.IntDenseNdArray;
+import org.tensorflow.ndarray.buffer.IntDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
+import org.tensorflow.types.family.TInteger;
 import org.tensorflow.types.family.TNumber;
 
-/** 32-bit signed integer tensor type. */
-public interface TInt32 extends IntNdArray, TNumber {
-
-  /** readable-name for the data type */
-  static final String NAME = "INT32";
-
-  /** Type metadata */
-  DataType<TInt32> DTYPE = DataType.create(NAME, 3, 4, TInt32Impl::mapTensor);
+/**
+ * 32-bit signed integer tensor type.
+ */
+@TensorType(dataType = DataType.DT_INT32, byteSize = 4, factory = TInt32Factory.class)
+public interface TInt32 extends TInteger, IntNdArray {
 
   /**
    * Allocates a new tensor for storing a single int value.
@@ -45,8 +42,8 @@ public interface TInt32 extends IntNdArray, TNumber {
    * @param value int to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TInt32> scalarOf(int value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setInt(value));
+  static TInt32 scalarOf(int value) {
+    return Tensors.of(TInt32.class, Shape.scalar(), t -> t.setInt(value));
   }
 
   /**
@@ -56,11 +53,11 @@ static Tensor<TInt32> scalarOf(int value) {
    * @return the new tensor
    * @throws IllegalArgumentException if no values are provided
    */
-  static Tensor<TInt32> vectorOf(int... values) {
+  static TInt32 vectorOf(int... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TInt32.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -71,8 +68,8 @@ static Tensor<TInt32> vectorOf(int... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TInt32> tensorOf(NdArray<Integer> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TInt32 tensorOf(NdArray<Integer> src) {
+    return Tensors.of(TInt32.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -81,8 +78,8 @@ static Tensor<TInt32> tensorOf(NdArray<Integer> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TInt32> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TInt32 tensorOf(Shape shape) {
+    return Tensors.of(TInt32.class, shape);
   }
 
   /**
@@ -92,30 +89,18 @@ static Tensor<TInt32> tensorOf(Shape shape) {
    * @param data buffer of ints to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TInt32> tensorOf(Shape shape, IntDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TInt32 tensorOf(Shape shape, IntDataBuffer data) {
+    return Tensors.of(TInt32.class, shape, t -> t.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    */
-  static Tensor<TInt32> tensorOf(Shape shape, Consumer<TInt32> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
-  }
-}
-
-/** Hidden implementation of a {@code TInt32} */
-class TInt32Impl extends IntDenseNdArray implements TInt32 {
-
-  static TInt32 mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TInt32Impl(TensorBuffers.toInts(nativeTensor), shape);
-  }
-
-  private TInt32Impl(IntDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
+  static TInt32 tensorOf(Shape shape, Consumer<TInt32> tensorInit) {
+    return Tensors.of(TInt32.class, shape, tensorInit);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt64.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt64.java
index 04fd4fd7799..cbefb248454 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt64.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TInt64.java
@@ -18,27 +18,24 @@
 package org.tensorflow.types;
 
 import java.util.function.Consumer;
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
+import org.tensorflow.Tensors;
 import org.tensorflow.exceptions.TensorFlowException;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.LongDataBuffer;
+import org.tensorflow.internal.types.TInt64Factory;
 import org.tensorflow.ndarray.LongNdArray;
 import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
-import org.tensorflow.ndarray.impl.dense.LongDenseNdArray;
+import org.tensorflow.ndarray.buffer.LongDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
+import org.tensorflow.types.family.TInteger;
 import org.tensorflow.types.family.TNumber;
 
-/** 64-bit signed integer tensor type. */
-public interface TInt64 extends LongNdArray, TNumber {
-
-  /** readable-name for the data type */
-  static final String NAME = "INT64";
-
-  /** Type metadata */
-  DataType<TInt64> DTYPE = DataType.create(NAME, 9, 8, TInt64Impl::mapTensor);
+/**
+ * 64-bit signed integer tensor type.
+ */
+@TensorType(dataType = DataType.DT_INT64, byteSize = 8, factory = TInt64Factory.class)
+public interface TInt64 extends TInteger, LongNdArray {
 
   /**
    * Allocates a new tensor for storing a single long value.
@@ -46,8 +43,8 @@ public interface TInt64 extends LongNdArray, TNumber {
    * @param value long to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TInt64> scalarOf(long value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setLong(value));
+  static TInt64 scalarOf(long value) {
+    return Tensors.of(TInt64.class, Shape.scalar(), t -> t.setLong(value));
   }
 
   /**
@@ -56,11 +53,11 @@ static Tensor<TInt64> scalarOf(long value) {
    * @param values longs to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TInt64> vectorOf(long... values) {
+  static TInt64 vectorOf(long... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TInt64.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -71,8 +68,8 @@ static Tensor<TInt64> vectorOf(long... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TInt64> tensorOf(NdArray<Long> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TInt64 tensorOf(NdArray<Long> src) {
+    return Tensors.of(TInt64.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -81,8 +78,8 @@ static Tensor<TInt64> tensorOf(NdArray<Long> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TInt64> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TInt64 tensorOf(Shape shape) {
+    return Tensors.of(TInt64.class, shape);
   }
 
   /**
@@ -92,31 +89,19 @@ static Tensor<TInt64> tensorOf(Shape shape) {
    * @param data buffer of longs to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TInt64> tensorOf(Shape shape, LongDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TInt64 tensorOf(Shape shape, LongDataBuffer data) {
+    return Tensors.of(TInt64.class, shape, t -> t.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    * @throws TensorFlowException if the tensor cannot be allocated or initialized
    */
-  static Tensor<TInt64> tensorOf(Shape shape, Consumer<TInt64> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
-  }
-}
-
-/** Hidden implementation of a {@code TInt64} */
-class TInt64Impl extends LongDenseNdArray implements TInt64 {
-
-  static TInt64 mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TInt64Impl(TensorBuffers.toLongs(nativeTensor), shape);
-  }
-
-  private TInt64Impl(LongDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
+  static TInt64 tensorOf(Shape shape, Consumer<TInt64> tensorInit) {
+    return Tensors.of(TInt64.class, shape, tensorInit);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TString.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TString.java
index 57a121edcf1..8c29ffb5aa6 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TString.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TString.java
@@ -17,24 +17,20 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
-import org.tensorflow.internal.buffer.StringTensorBuffer;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
+import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
+import java.util.function.Function;
+import org.tensorflow.Tensors;
+import org.tensorflow.internal.buffer.ByteSequenceTensorBuffer;
+import org.tensorflow.internal.types.TStringFactory;
 import org.tensorflow.ndarray.NdArray;
 import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.buffer.DataBuffer;
-import org.tensorflow.ndarray.buffer.layout.DataLayout;
-import org.tensorflow.ndarray.buffer.layout.DataLayouts;
-import org.tensorflow.ndarray.impl.dense.DenseNdArray;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
 import org.tensorflow.types.family.TType;
 
-import java.nio.charset.Charset;
-import java.nio.charset.StandardCharsets;
-import java.util.function.Function;
-
 /**
  * String type.
  *
@@ -44,13 +40,8 @@
  * its values initially, so TensorFlow can compute and allocate the right amount of memory. Then the
  * data in the tensor is initialized once and cannot be modified afterwards.
  */
-public interface TString extends NdArray<String>, TType {
-
-  /** readable-name for the data type */
-  static final String NAME = "STRING";
-
-  /** Type metadata */
-  DataType<TString> DTYPE = DataType.create(NAME, 7, -1, TStringImpl::mapTensor);
+@TensorType(dataType = DataType.DT_STRING, byteSize = -1, factory = TStringFactory.class)
+public interface TString extends TType, NdArray<String> {
 
   /**
    * Allocates a new tensor for storing a string scalar.
@@ -60,7 +51,7 @@ public interface TString extends NdArray<String>, TType {
    * @param value scalar value to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TString> scalarOf(String value) {
+  static TString scalarOf(String value) {
     return tensorOf(NdArrays.scalarOfObject(value));
   }
 
@@ -72,7 +63,7 @@ static Tensor<TString> scalarOf(String value) {
    * @param values values to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TString> vectorOf(String... values) {
+  static TString vectorOf(String... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
@@ -88,7 +79,7 @@ static Tensor<TString> vectorOf(String... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TString> tensorOf(NdArray<String> src) {
+  static TString tensorOf(NdArray<String> src) {
     return tensorOf(StandardCharsets.UTF_8, src);
   }
 
@@ -113,8 +104,10 @@ static Tensor<TString> tensorOf(NdArray<String> src) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TString> tensorOf(Charset charset, NdArray<String> src) {
-    return TStringImpl.createTensor(src, s -> s.getBytes(charset));
+  static TString tensorOf(Charset charset, NdArray<String> src) {
+    Function<String, byte[]> getBytes = s -> s.getBytes(charset);
+    long size = ByteSequenceTensorBuffer.computeSize(src, getBytes);
+    return Tensors.of(TString.class, src.shape(), size, t -> t.write(src, getBytes));
   }
 
   /**
@@ -127,7 +120,7 @@ static Tensor<TString> tensorOf(Charset charset, NdArray<String> src) {
    * @param data buffer of strings to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TString> tensorOf(Shape shape, DataBuffer<String> data) {
+  static TString tensorOf(Shape shape, DataBuffer<String> data) {
     return tensorOf(NdArrays.wrap(shape, data));
   }
 
@@ -154,7 +147,7 @@ static Tensor<TString> tensorOf(Shape shape, DataBuffer<String> data) {
    * @param data buffer of strings to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TString> tensorOf(Charset charset, Shape shape, DataBuffer<String> data) {
+  static TString tensorOf(Charset charset, Shape shape, DataBuffer<String> data) {
     return tensorOf(charset, NdArrays.wrap(shape, data));
   }
 
@@ -173,8 +166,10 @@ static Tensor<TString> tensorOf(Charset charset, Shape shape, DataBuffer<String>
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TString> tensorOfBytes(NdArray<byte[]> src) {
-    return TStringImpl.createTensor(src, Function.identity());
+  static TString tensorOfBytes(NdArray<byte[]> src) {
+    Function<byte[], byte[]> getBytes = Function.identity();
+    long size = ByteSequenceTensorBuffer.computeSize(src, getBytes);
+    return Tensors.of(TString.class, src.shape(), size, t -> t.write(src, getBytes));
   }
 
   /**
@@ -193,7 +188,7 @@ static Tensor<TString> tensorOfBytes(NdArray<byte[]> src) {
    * @param data the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TString> tensorOfBytes(Shape shape, DataBuffer<byte[]> data) {
+  static TString tensorOfBytes(Shape shape, DataBuffer<byte[]> data) {
     return tensorOfBytes(NdArrays.wrap(shape, data));
   }
 
@@ -215,45 +210,7 @@ static Tensor<TString> tensorOfBytes(Shape shape, DataBuffer<byte[]> data) {
    */
   TString using(Charset charset);
 
-  /** @return the tensor data as a n-dimensional array of raw byte sequences. */
-  NdArray<byte[]> asBytes();
-}
-
-/** Hidden implementation of a {@code TString} */
-class TStringImpl extends DenseNdArray<String> implements TString {
+  <T> void write(NdArray<T> src, Function<T, byte[]> getBytes);
 
-  @Override
-  public TString using(Charset charset) {
-    return new TStringImpl(tensorBuffer, DataLayouts.ofStrings(charset), shape());
-  }
-
-  @Override
-  public NdArray<byte[]> asBytes() {
-    return NdArrays.wrap(shape(), tensorBuffer);
-  }
-
-  static <T> Tensor<TString> createTensor(NdArray<T> src, Function<T, byte[]> getBytes) {
-    long size = StringTensorBuffer.computeSize(src, getBytes);
-    return Tensor.of(
-        TString.DTYPE,
-        src.shape(),
-        size,
-        data -> ((TStringImpl) data).tensorBuffer.init(src, getBytes));
-  }
-
-  static TString mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    StringTensorBuffer buffer = TensorBuffers.toStrings(nativeTensor, shape.size());
-    return new TStringImpl(buffer, UTF_8_LAYOUT, shape);
-  }
-
-  private static DataLayout<DataBuffer<byte[]>, String> UTF_8_LAYOUT =
-      DataLayouts.ofStrings(StandardCharsets.UTF_8);
-
-  private final StringTensorBuffer tensorBuffer;
-
-  private TStringImpl(
-      StringTensorBuffer buffer, DataLayout<DataBuffer<byte[]>, String> layout, Shape shape) {
-    super(layout.applyTo(buffer), shape);
-    tensorBuffer = buffer;
-  }
+  NdArray<byte[]> asBytes();
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TUint8.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TUint8.java
index 365f41196fb..c92a2db560e 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TUint8.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TUint8.java
@@ -18,27 +18,24 @@
 package org.tensorflow.types;
 
 import java.util.function.Consumer;
-import org.tensorflow.DataType;
-import org.tensorflow.Tensor;
+import org.tensorflow.Tensors;
 import org.tensorflow.exceptions.TensorFlowException;
-import org.tensorflow.internal.buffer.TensorBuffers;
-import org.tensorflow.internal.c_api.TF_Tensor;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.ByteDataBuffer;
+import org.tensorflow.internal.types.TUint8Factory;
 import org.tensorflow.ndarray.ByteNdArray;
 import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
-import org.tensorflow.ndarray.impl.dense.ByteDenseNdArray;
+import org.tensorflow.ndarray.buffer.ByteDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
+import org.tensorflow.types.family.TInteger;
 import org.tensorflow.types.family.TNumber;
 
-/** 8-bit unsigned integer tensor type. */
-public interface TUint8 extends ByteNdArray, TNumber {
-
-  /** readable-name for the data type */
-  static final String NAME = "UINT8";
-
-  /** Type metadata */
-  DataType<TUint8> DTYPE = DataType.create(NAME, 4, 1, TUint8Impl::mapTensor);
+/**
+ * 8-bit unsigned integer tensor type.
+ */
+@TensorType(dataType = DataType.DT_UINT8, byteSize = 1, factory = TUint8Factory.class)
+public interface TUint8 extends TInteger, ByteNdArray {
 
   /**
    * Allocates a new tensor for storing a single byte value.
@@ -46,8 +43,8 @@ public interface TUint8 extends ByteNdArray, TNumber {
    * @param value byte to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TUint8> scalarOf(byte value) {
-    return Tensor.of(DTYPE, Shape.scalar(), data -> data.setByte(value));
+  static TUint8 scalarOf(byte value) {
+    return Tensors.of(TUint8.class, Shape.scalar(), t -> t.setByte(value));
   }
 
   /**
@@ -56,11 +53,11 @@ static Tensor<TUint8> scalarOf(byte value) {
    * @param values bytes to store in the new tensor
    * @return the new tensor
    */
-  static Tensor<TUint8> vectorOf(byte... values) {
+  static TUint8 vectorOf(byte... values) {
     if (values == null) {
       throw new IllegalArgumentException();
     }
-    return Tensor.of(DTYPE, Shape.of(values.length), data -> StdArrays.copyTo(values, data));
+    return Tensors.of(TUint8.class, Shape.of(values.length), t -> StdArrays.copyTo(values, t));
   }
 
   /**
@@ -71,8 +68,8 @@ static Tensor<TUint8> vectorOf(byte... values) {
    * @param src the source array giving the shape and data to the new tensor
    * @return the new tensor
    */
-  static Tensor<TUint8> tensorOf(NdArray<Byte> src) {
-    return Tensor.of(DTYPE, src.shape(), src::copyTo);
+  static TUint8 tensorOf(NdArray<Byte> src) {
+    return Tensors.of(TUint8.class, src.shape(), src::copyTo);
   }
 
   /**
@@ -81,8 +78,8 @@ static Tensor<TUint8> tensorOf(NdArray<Byte> src) {
    * @param shape shape of the tensor to allocate
    * @return the new tensor
    */
-  static Tensor<TUint8> tensorOf(Shape shape) {
-    return Tensor.of(DTYPE, shape);
+  static TUint8 tensorOf(Shape shape) {
+    return Tensors.of(TUint8.class, shape);
   }
 
   /**
@@ -92,31 +89,19 @@ static Tensor<TUint8> tensorOf(Shape shape) {
    * @param data buffer of bytes to initialize the tensor with
    * @return the new tensor
    */
-  static Tensor<TUint8> tensorOf(Shape shape, ByteDataBuffer data) {
-    return Tensor.of(DTYPE, shape, d -> d.write(data));
+  static TUint8 tensorOf(Shape shape, ByteDataBuffer data) {
+    return Tensors.of(TUint8.class, shape, d -> d.write(data));
   }
 
   /**
    * Allocates a new tensor of the given shape and initialize its data.
    *
    * @param shape shape of the tensor to allocate
-   * @param dataInit tensor data initializer
+   * @param tensorInit tensor data initializer
    * @return the new tensor
    * @throws TensorFlowException if the tensor cannot be allocated or initialized
    */
-  static Tensor<TUint8> tensorOf(Shape shape, Consumer<TUint8> dataInit) {
-    return Tensor.of(DTYPE, shape, dataInit);
-  }
-}
-
-/** Hidden implementation of a {@code TUint8} */
-class TUint8Impl extends ByteDenseNdArray implements TUint8 {
-
-  static TUint8 mapTensor(TF_Tensor nativeTensor, Shape shape) {
-    return new TUint8Impl(TensorBuffers.toBytes(nativeTensor), shape);
-  }
-
-  private TUint8Impl(ByteDataBuffer buffer, Shape shape) {
-    super(buffer, shape);
+  static TUint8 tensorOf(Shape shape, Consumer<TUint8> tensorInit) {
+    return Tensors.of(TUint8.class, shape, tensorInit);
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/Type.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/Type.java
new file mode 100644
index 00000000000..499f393d7cc
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/Type.java
@@ -0,0 +1,39 @@
+package org.tensorflow.types;
+
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.family.TType;
+
+public class Type<T extends TType> {
+
+  public Class<T> typeClass() {
+    return typeClass;
+  }
+
+  public DataType dataType() {
+    return dataType;
+  }
+
+  public int byteSize() {
+    return byteSize;
+  }
+
+  public boolean isVariableLength() {
+    return byteSize < 0;
+  }
+
+  public TypeFactory<T> factory() {
+    return factory;
+  }
+
+  Type(Class<T> typeClass, DataType dataType, int byteSize, TypeFactory<T> factory) {
+    this.typeClass = typeClass;
+    this.dataType = dataType;
+    this.byteSize = byteSize;
+    this.factory = factory;
+  }
+
+  private final Class<T> typeClass;
+  private final DataType dataType;
+  private final int byteSize;
+  private final TypeFactory<T> factory;
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TypeFactory.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TypeFactory.java
new file mode 100644
index 00000000000..4529657003e
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TypeFactory.java
@@ -0,0 +1,10 @@
+package org.tensorflow.types;
+
+import org.tensorflow.TensorHandle;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.types.family.TType;
+
+public interface TypeFactory<T extends TType> {
+
+  T createDense(TensorHandle tensorHandle, Shape shape);
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TypeRegistry.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TypeRegistry.java
new file mode 100644
index 00000000000..99e9170fa55
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/TypeRegistry.java
@@ -0,0 +1,98 @@
+/*
+ *  Copyright 2019 The TensorFlow Authors. All Rights Reserved.
+ *
+ *  Licensed under the Apache License, Version 2.0 (the "License");
+ *  you may not use this file except in compliance with the License.
+ *  You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ *  Unless required by applicable law or agreed to in writing, software
+ *  distributed under the License is distributed on an "AS IS" BASIS,
+ *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ *  See the License for the specific language governing permissions and
+ *  limitations under the License.
+ *  =======================================================================
+ */
+
+package org.tensorflow.types;
+
+import java.util.HashMap;
+import java.util.Map;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.annotation.TensorType;
+import org.tensorflow.types.family.TType;
+
+/**
+ * Utility class for working with {@link DataType} objects.
+ */
+public final class TypeRegistry {
+
+  /**
+   * Find a type registration from a data type
+   *
+   * @param dataType data type
+   * @return type registered to this code
+   * @throws IllegalArgumentException if the code matches no registered data type
+   */
+  public static <T extends TType> Type<T> find(DataType dataType) {
+    Type<?> entry = TYPES_BY_CODE.get(dataType.getNumber());
+    if (entry == null) {
+      throw new IllegalArgumentException("No type has been registered for datatype " + dataType);
+    }
+    return (Type<T>)entry;
+  }
+
+  /**
+   * Find a type registration from a type class
+   *
+   * @param typeClass class implementing {@link Tensor}
+   * @return type registration
+   * @throws IllegalArgumentException if the code matches no registered data type
+   */
+  public static <T extends TType> Type<T> find(Class<T> typeClass) {
+    Type<?> entry = TYPES_BY_CLASS.get(typeClass);
+    if (entry == null) {
+      throw new IllegalArgumentException("Class \"" + typeClass.getName() + "\" is not a valid datatype class");
+    }
+    return (Type<T>)entry;
+  }
+
+  private static final Map<Integer, Type<?>> TYPES_BY_CODE = new HashMap<>();
+  private static final Map<Class<? extends TType>, Type<?>> TYPES_BY_CLASS = new HashMap<>();
+
+  private static <T extends TType> void register(Class<T> typeClass) {
+    TensorType typeAnnot = typeClass.getDeclaredAnnotation(TensorType.class);
+    if (typeAnnot == null) {
+      throw new IllegalArgumentException("Class \"" + typeClass.getName() + "\" must be annotated "
+          + "with @TensorType to be registered as a tensor type");
+    }
+    TypeFactory<T> factory;
+    try {
+      factory = typeAnnot.factory().newInstance();
+    } catch (ReflectiveOperationException e) {
+      throw new IllegalArgumentException("Class \"" + typeClass.getName() + "\" must have a public "
+          + "parameter-less constructor to be used as a tensor type factory");
+    }
+    Type<T> type = new Type<T>(typeClass, typeAnnot.dataType(), typeAnnot.byteSize(), factory);
+    TYPES_BY_CLASS.put(typeClass, type);
+
+    // If more than one tensor type is mapped to a given native code, the last registered will
+    // have priority. This way, we can allow user to register their own classes to map tensors
+    // of a given data type.
+    TYPES_BY_CODE.put(type.dataType().getNumber(), type);
+    TYPES_BY_CODE.put(type.dataType().getNumber() + 100, type);
+  }
+
+  static {
+    register(TBool.class);
+    register(TFloat64.class);
+    register(TFloat32.class);
+    register(TFloat16.class);
+    register(TInt32.class);
+    register(TInt64.class);
+    register(TString.class);
+    register(TUint8.class);
+    register(TBfloat16.class);
+  }
+}
\ No newline at end of file
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/annotation/TensorType.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/annotation/TensorType.java
new file mode 100644
index 00000000000..b5da7ddcb35
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/annotation/TensorType.java
@@ -0,0 +1,40 @@
+/* Copyright 2016 The TensorFlow Authors. All Rights Reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+==============================================================================*/
+
+package org.tensorflow.types.annotation;
+
+import java.lang.annotation.Documented;
+import java.lang.annotation.ElementType;
+import java.lang.annotation.Retention;
+import java.lang.annotation.RetentionPolicy;
+import java.lang.annotation.Target;
+import org.tensorflow.proto.framework.DataType;
+import org.tensorflow.types.TypeFactory;
+
+/** Represents a type of elements in a {@link Tensor} */
+@Documented
+@Retention(RetentionPolicy.RUNTIME)
+@Target(ElementType.TYPE)
+public @interface TensorType {
+
+  DataType dataType();
+
+  int byteSize();
+
+  /**
+   * The class implementing this tensor type
+   */
+  Class<? extends TypeFactory> factory();
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/TInteger.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/TInteger.java
new file mode 100644
index 00000000000..0112d278e49
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/TInteger.java
@@ -0,0 +1,5 @@
+package org.tensorflow.types.family;
+
+public interface TInteger extends TNumber {
+
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/TType.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/TType.java
index 8f3451b9a68..8202d22f570 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/TType.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/TType.java
@@ -1,37 +1,31 @@
-/*
- *  Copyright 2019 The TensorFlow Authors. All Rights Reserved.
- *
- *  Licensed under the Apache License, Version 2.0 (the "License");
- *  you may not use this file except in compliance with the License.
- *  You may obtain a copy of the License at
- *
- *      http://www.apache.org/licenses/LICENSE-2.0
- *
- *  Unless required by applicable law or agreed to in writing, software
- *  distributed under the License is distributed on an "AS IS" BASIS,
- *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- *  See the License for the specific language governing permissions and
- *  limitations under the License.
- *  =======================================================================
- */
-
 package org.tensorflow.types.family;
 
-/**
- * Marker interface for all tensor types.
- *
- * <p>Tensor types are carried as a generic parameter of the {@link org.tensorflow.Tensor Tensor}
- * class bound by the {@code TType} interface. This generic parameter ensure type-compatibility
- * between operands of a computation at compile-time. For example:
- *
- * <pre>{@code
- * Tensor<TFloat32> tensor1 = TFloat32.ofShape(2, 3, 2);
- * Tensor<TFloat32> tensor2 = TFloat32.ofShape(2, 3, 2);
- * Tensor<TInt32> tensor3 = TInt32.ofShape(2, 3, 2);
- *
- * Ops tf = Ops.create();
- * tf.math.add(tf.constant(tensor1), tf.constant(tensor2));  // OK
- * tf.math.add(tf.constant(tensor1), tf.constant(tensor3));  // Compilation failure
- * }</pre>
- */
-public interface TType {}
+import org.tensorflow.Tensor;
+import org.tensorflow.TensorHandle;
+import org.tensorflow.ndarray.buffer.ByteDataBuffer;
+import org.tensorflow.proto.framework.DataType;
+
+public interface TType extends Tensor {
+
+  Class<? extends TType> type();
+
+  @Override
+  default void close() {
+    handle().close();
+  }
+
+  @Override
+  default ByteDataBuffer rawData() {
+    return handle().rawData();
+  }
+
+  @Override
+  default DataType dataType() {
+    return handle().dataType();
+  }
+
+  @Override
+  default long numBytes() {
+    return handle().numBytes();
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/package-info.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/package-info.java
index fcf1f554133..79e9e93fea3 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/package-info.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/family/package-info.java
@@ -21,8 +21,5 @@
  * <p>Some operations enforces that only operands of a type from a given family can be passed
  * in argument. For example, if an operation only allows numeric operands, such operands must be
  * bound to the {@link org.tensorflow.types.family.TNumber TNumber} interface.
- *
- * <p>All tensor types is bound to {@link org.tensorflow.types.family.TType TType}, which lays at
- * the root of the family hierarchy.
  */
 package org.tensorflow.types.family;
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/package-info.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/package-info.java
index afbd69fabe5..b41ce514f00 100644
--- a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/package-info.java
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/types/package-info.java
@@ -32,10 +32,10 @@
  *
  * <p>Instances of tensor types must also implement the {@link org.tensorflow.ndarray.NdArray NdArray}
  * interface so a user can access directly the tensor data in a n-dimensional space by invoking
- * {@link org.tensorflow.Tensor#data() Tensor.data()}.
+ * {@link org.tensorflow.util.Tensor#data() Tensor.data()}.
  *
  * <p>Note that while it is always possible to allocate a tensor using the
- * {@link org.tensorflow.Tensor#of(org.tensorflow.DataType, Shape) Tensor.of(...)}
+ * {@link org.tensorflow.util.Tensor#of(org.tensorflow.DataType, Shape) Tensor.of(...)}
  * method, most tensor types expose factory methods that simplify the creation process, like
  * {@code scalarOf(...)}, {@code vectorOf(...)}, {@code tensorOf(...)}, etc.
  */
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/util/TensorList.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/util/TensorList.java
new file mode 100644
index 00000000000..87a4fb378c9
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/util/TensorList.java
@@ -0,0 +1,60 @@
+package org.tensorflow.util;
+
+import java.util.ArrayList;
+import java.util.Collection;
+import java.util.Iterator;
+import java.util.List;
+import java.util.ListIterator;
+import java.util.stream.Stream;
+import org.tensorflow.types.family.TType;
+
+public class TensorList implements AutoCloseable {
+
+  public boolean isEmpty() {
+    return tensors.isEmpty();
+  }
+
+  public int size() {
+    return tensors.size();
+  }
+
+  public Iterator<? extends TType> iterator() {
+    return tensors.iterator();
+  }
+
+  public ListIterator<? extends TType> listIterator() {
+    return tensors.listIterator();
+  }
+
+  public Stream<? extends TType> stream() {
+    return tensors.stream();
+  }
+
+  public <T extends TType> boolean add(T tensor) {
+    return tensors.add(tensor);
+  }
+
+  public <T extends TType> T single() {
+    if (tensors.size() != 1) {
+      throw new IllegalStateException("List must contain a single tensor to use non-indexed getter");
+    }
+    return get(0);
+  }
+
+  public <T extends TType> T get(int index) {
+    return (T)tensors.get(index);
+  }
+
+  public TensorMap toMap(Collection<String> tensorNames) {
+    return new TensorMap(tensorNames, this);
+  }
+
+  @Override
+  public void close() {
+    for (TType t : tensors) {
+      t.handle().release();
+    }
+  }
+
+  private final List<TType> tensors = new ArrayList<>();
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/util/TensorMap.java b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/util/TensorMap.java
new file mode 100644
index 00000000000..7e821ccf2af
--- /dev/null
+++ b/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/util/TensorMap.java
@@ -0,0 +1,72 @@
+package org.tensorflow.util;
+
+import java.util.Collection;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+import org.tensorflow.types.family.TType;
+
+public class TensorMap extends HashMap<String, TType> implements AutoCloseable {
+
+  @Override
+  public TType put(String key, TType value) {
+    value.handle().retain();
+    TType t = super.put(key, value);
+    if (t != null) {
+      t.handle().release();
+    }
+    return t;
+  }
+
+  @Override
+  public TType remove(Object key) {
+    TType t = super.remove(key);
+    if (t != null) {
+      t.handle().release();
+    }
+    return t;
+  }
+
+  @Override
+  public void putAll(Map<? extends String, ? extends TType> m) {
+    super.putAll(m);
+    for (TType t : m.values()) {
+      t.handle().retain();
+    }
+  }
+
+  @Override
+  public void clear() {
+    for (TType t : values()) {
+      t.handle().release();
+    }
+    super.clear();
+  }
+
+  @Override
+  public void close() {
+    clear();
+  }
+
+  public <T extends TType> T get(String name) {
+    return (T) super.get(name);
+  }
+
+  public <T extends TType> T single() {
+    if (size() != 1) {
+      throw new IllegalStateException("List must contain a single tensor to use non-indexed getter");
+    }
+    return (T) values().iterator().next();
+  }
+
+  public TensorMap() {
+    super();
+  }
+
+  public TensorMap(Collection<String> names, TensorList tensors) {
+    Iterator<? extends TType> tensorIter = tensors.iterator();
+    for (String name : names) {
+      put(name, tensorIter.next());
+    }
+  }
+}
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/AutoCloseableList.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/AutoCloseableList.java
deleted file mode 100644
index 330a40bae6b..00000000000
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/AutoCloseableList.java
+++ /dev/null
@@ -1,27 +0,0 @@
-package org.tensorflow;
-
-import java.util.ArrayList;
-import java.util.Collection;
-
-public final class AutoCloseableList<E extends AutoCloseable> extends ArrayList<E>
-    implements AutoCloseable {
-
-  public AutoCloseableList(Collection<? extends E> c) {
-    super(c);
-  }
-
-  @Override
-  public void close() {
-    Exception toThrow = null;
-    for (AutoCloseable c : this) {
-      try {
-        c.close();
-      } catch (Exception e) {
-        toThrow = e;
-      }
-    }
-    if (toThrow != null) {
-      throw new RuntimeException(toThrow);
-    }
-  }
-}
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/ConcreteFunctionTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/ConcreteFunctionTest.java
index 3ea20fcbb46..dd17c891e69 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/ConcreteFunctionTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/ConcreteFunctionTest.java
@@ -29,14 +29,14 @@
 public class ConcreteFunctionTest {
 
   private static Signature plusFive(Ops tf) {
-    Placeholder<TFloat32> input = tf.placeholder(TFloat32.DTYPE);
+    Placeholder<TFloat32> input = tf.placeholder(TFloat32.class);
     Add<TFloat32> output = tf.math.add(input, tf.constant(5.0f));
     Init init = tf.init();  // for native resource management tests
     return Signature.builder().key("plusFive").input("x", input).output("y", output).build();
   }
 
   private static Signature minusTwo(Ops tf) {
-    Placeholder<TFloat32> input = tf.placeholder(TFloat32.DTYPE);
+    Placeholder<TFloat32> input = tf.placeholder(TFloat32.class);
     Sub<TFloat32> output = tf.math.sub(input, tf.constant(2.0f));
     return Signature.builder().key("minusTwo").input("x", input).output("y", output).build();
   }
@@ -44,8 +44,8 @@ private static Signature minusTwo(Ops tf) {
   @Test
   public void createFunction() {
     try (ConcreteFunction f = ConcreteFunction.create(ConcreteFunctionTest::plusFive);
-        Tensor<TFloat32> x = TFloat32.scalarOf(3.0f)) {
-      assertEquals(8.0f, f.call(x).expect(TFloat32.DTYPE).data().getFloat());
+        TFloat32 x = TFloat32.scalarOf(3.0f)) {
+      assertEquals(8.0f, ((TFloat32)f.call(x)).getFloat());
     }
   }
 
@@ -54,8 +54,8 @@ public void createFunctionFromGraph() {
     try (Graph g = new Graph()) {
       Signature signature = plusFive(Ops.create(g));
       try (ConcreteFunction f = ConcreteFunction.create(signature, g);
-          Tensor<TFloat32> x = TFloat32.scalarOf(3.0f)) {
-        assertEquals(8.0f, f.call(x).expect(TFloat32.DTYPE).data().getFloat());
+          TFloat32 x = TFloat32.scalarOf(3.0f)) {
+        assertEquals(8.0f, ((TFloat32)f.call(x)).getFloat());
       }
     }
   }
@@ -66,8 +66,8 @@ public void createFunctionFromSession() {
       Signature signature = plusFive(Ops.create(g));
       try (Session s = new Session(g)) {
         try (ConcreteFunction f = ConcreteFunction.create(signature, s);
-            Tensor<TFloat32> x = TFloat32.scalarOf(3.0f)) {
-          assertEquals(8.0f, f.call(x).expect(TFloat32.DTYPE).data().getFloat());
+            TFloat32 x = TFloat32.scalarOf(3.0f)) {
+          assertEquals(8.0f, ((TFloat32)f.call(x)).getFloat());
         }
       }
     }
@@ -77,8 +77,9 @@ public void createFunctionFromSession() {
   public void chainFunctions() {
     try (ConcreteFunction f1 = ConcreteFunction.create(ConcreteFunctionTest::plusFive);
         ConcreteFunction f2 = ConcreteFunction.create(ConcreteFunctionTest::minusTwo);
-        Tensor<TFloat32> x = TFloat32.scalarOf(3.0f)) {
-      assertEquals(6.0f, f2.call(f1.call(x)).expect(TFloat32.DTYPE).data().getFloat());
+        TFloat32 x = TFloat32.scalarOf(3.0f)) {
+      TFloat32 y = f1.call(x);
+      assertEquals(6.0f, ((TFloat32)f2.call(y)).getFloat());
     }
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationBuilderTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationBuilderTest.java
index 6802ead9592..fda6084b796 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationBuilderTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationBuilderTest.java
@@ -45,7 +45,7 @@ public void failToBuildOpIfSessionIsClosed() {
       opBuilder = new EagerOperationBuilder(session, "Empty", "empty");
     }
     try {
-      opBuilder.setAttr("dtype", TFloat32.DTYPE);
+      opBuilder.setAttr("dtype", TFloat32.class);
       fail();
     } catch (IllegalStateException e) {
       // expected
@@ -93,9 +93,9 @@ public void setAttrs() {
     try (EagerSession session = EagerSession.create()) {
       Ops tf = Ops.create(session);
       // dtype, tensor attributes.
-      try (Tensor<TInt32> t = TInt32.scalarOf(1)) {
+      try (TInt32 t = TInt32.scalarOf(1)) {
         opBuilder(session, "Const", "DataTypeAndTensor")
-            .setAttr("dtype", TInt32.DTYPE)
+            .setAttr("dtype", TInt32.class)
             .setAttr("value", t)
             .build();
       }
@@ -103,7 +103,7 @@ public void setAttrs() {
       opBuilder(session, "RandomUniform", "DataTypeAndInt")
           .addInput(tf.array(1).asOutput())
           .setAttr("seed", 10)
-          .setAttr("dtype", TFloat32.DTYPE)
+          .setAttr("dtype", TFloat32.class)
           .build();
       // list(int), string
       opBuilder(session, "MaxPool", "IntListAndString")
@@ -124,7 +124,7 @@ public void setAttrs() {
           .build();
       // list(shape)
       opBuilder(session, "FIFOQueue", "queue")
-          .setAttr("component_types", new DataType[] {TInt32.DTYPE, TInt32.DTYPE})
+          .setAttr("component_types", new Class[] {TInt32.class, TInt32.class})
           .setAttr("shapes", new Shape[] {Shape.of(2, 2), Shape.of(2, 2, 2)})
           .build();
       // bool
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationTest.java
index 09d2214cc6a..c702f3e9883 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/EagerOperationTest.java
@@ -23,6 +23,7 @@
 import org.tensorflow.exceptions.TFInvalidArgumentException;
 import org.tensorflow.op.Ops;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.proto.framework.DataType;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 
@@ -46,13 +47,13 @@ public void failToCreateIfSessionIsClosed() {
   @Test
   public void outputDataTypeAndShape() {
     try (EagerSession session = EagerSession.create();
-         Tensor<TInt32> t = TInt32.tensorOf(Shape.of(2, 3))) {
+         TInt32 t = TInt32.tensorOf(Shape.of(2, 3))) {
       EagerOperation op =
           opBuilder(session, "Const", "OutputAttrs")
-              .setAttr("dtype", TInt32.DTYPE)
+              .setAttr("dtype", TInt32.class)
               .setAttr("value", t)
               .build();
-      assertEquals(TInt32.DTYPE, op.dtype(0));
+      assertEquals(DataType.DT_INT32, op.dtype(0));
       assertEquals(2, op.shape(0).size(0));
       assertEquals(3, op.shape(0).size(1));
     }
@@ -67,12 +68,12 @@ public void outputTensor() {
               .addInput(tf.constant(2).asOutput())
               .addInput(tf.constant(4).asOutput())
               .build();
-      assertEquals(6, add.tensor(0).expect(TInt32.DTYPE).data().getInt());
+      assertEquals(6, ((TInt32)Tensors.fromHandle(add.tensor(0))).getInt());
 
       // Validate that we retrieve the right shape and datatype from the tensor
       // that has been resolved
       assertEquals(0, add.shape(0).numDimensions());
-      assertEquals(TInt32.DTYPE, add.dtype(0));
+      assertEquals(DataType.DT_INT32, add.dtype(0));
     }
   }
 
@@ -123,7 +124,7 @@ public void numOutputs() {
           opBuilder(session, "UniqueWithCountsV2", "unq")
               .addInput(tf.constant(new int[]{1, 2, 1}).asOutput())
               .addInput(tf.constant(new int[]{0}).asOutput())
-              .setAttr("out_idx", TInt32.DTYPE)
+              .setAttr("out_idx", TInt32.class)
               .build();
       assertEquals(3, op.numOutputs());
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphOperationBuilderTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphOperationBuilderTest.java
index 35bfa808238..4d05dd9ad26 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphOperationBuilderTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphOperationBuilderTest.java
@@ -50,12 +50,12 @@ public void failWhenMixingOperationsOnDifferentGraphs() {
   @Test
   public void failOnUseAfterBuild() {
     try (Graph g = new Graph();
-        Tensor<TInt32> t = TInt32.scalarOf(1)) {
+        TInt32 t = TInt32.scalarOf(1)) {
       OperationBuilder b =
-          g.opBuilder("Const", "Const").setAttr("dtype", t.dataType()).setAttr("value", t);
+          g.opBuilder("Const", "Const").setAttr("dtype", t.type()).setAttr("value", t);
       b.build();
       try {
-        b.setAttr("dtype", t.dataType());
+        b.setAttr("dtype", t.type());
       } catch (IllegalStateException e) {
         // expected exception.
       }
@@ -66,8 +66,8 @@ public void failOnUseAfterBuild() {
   public void failOnUseAfterGraphClose() {
     OperationBuilder b = null;
     try (Graph g = new Graph();
-        Tensor<TInt32> t = TInt32.scalarOf(1)) {
-      b = g.opBuilder("Const", "Const").setAttr("dtype", t.dataType()).setAttr("value", t);
+        TInt32 t = TInt32.scalarOf(1)) {
+      b = g.opBuilder("Const", "Const").setAttr("dtype", t.type()).setAttr("value", t);
     }
     try {
       b.build();
@@ -88,9 +88,9 @@ public void setAttr() {
     try (Graph g = new Graph()) {
       Ops tf = Ops.create(g);
       // dtype, tensor attributes.
-      try (Tensor<TInt32> t = TInt32.scalarOf(1)) {
+      try (TInt32 t = TInt32.scalarOf(1)) {
         g.opBuilder("Const", "DataTypeAndTensor")
-            .setAttr("dtype", TInt32.DTYPE)
+            .setAttr("dtype", TInt32.class)
             .setAttr("value", t)
             .build()
             .output(0);
@@ -106,7 +106,7 @@ public void setAttr() {
       g.opBuilder("RandomUniform", "Int")
           .addInput(tf.array(1).asOutput())
           .setAttr("seed", 10)
-          .setAttr("dtype", TFloat32.DTYPE)
+          .setAttr("dtype", TFloat32.class)
           .build();
       assertTrue(hasNode(g, "Int"));
       // list(int)
@@ -132,23 +132,23 @@ public void setAttrShape() {
     try (Graph g = new Graph()) {
       Output<?> n =
           g.opBuilder("Placeholder", "unknown")
-              .setAttr("dtype", TFloat32.DTYPE)
+              .setAttr("dtype", TFloat32.class)
               .setAttr("shape", Shape.unknown())
               .build()
               .output(0);
       assertEquals(-1, n.shape().numDimensions());
-      assertEquals(TFloat32.DTYPE, n.dataType());
+      assertEquals(TFloat32.class, n.type());
 
       n =
           g.opBuilder("Placeholder", "batch_of_vectors")
-              .setAttr("dtype", TFloat32.DTYPE)
+              .setAttr("dtype", TFloat32.class)
               .setAttr("shape", Shape.of(-1, 784))
               .build()
               .output(0);
       assertEquals(2, n.shape().numDimensions());
       assertEquals(-1, n.shape().size(0));
       assertEquals(784, n.shape().size(1));
-      assertEquals(TFloat32.DTYPE, n.dataType());
+      assertEquals(TFloat32.class, n.type());
     }
   }
 
@@ -169,10 +169,10 @@ public void setAttrShapeList() {
   public void addControlInput() {
     try (Graph g = new Graph();
         Session s = new Session(g);
-        Tensor<TBool> yes = TBool.scalarOf(true);
-        Tensor<TBool> no = TBool.scalarOf(false)) {
+        TBool yes = TBool.scalarOf(true);
+        TBool no = TBool.scalarOf(false)) {
       Ops tf = Ops.create(g);
-      Output<TBool> placeholder = tf.placeholder(TBool.DTYPE).asOutput();
+      Output<TBool> placeholder = tf.placeholder(TBool.class).asOutput();
       GraphOperation check =
           g.opBuilder("Assert", "assert")
               .addInput(placeholder)
@@ -200,7 +200,7 @@ private static void testSetAttrShapeList(Shape[] shapes) {
       int[][] matrix = new int[][] {{0, 0}, {0, 0}};
       Output<?> queue =
           g.opBuilder("FIFOQueue", "queue")
-              .setAttr("component_types", new DataType[] {TInt32.DTYPE, TInt32.DTYPE})
+              .setAttr("component_types", new Class[] {TInt32.class, TInt32.class})
               .setAttr("shapes", shapes)
               .build()
               .output(0);
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphTest.java
index de376015e3f..64643c0c58c 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/GraphTest.java
@@ -28,6 +28,7 @@
 import org.tensorflow.op.Ops;
 import org.tensorflow.op.linalg.MatMul;
 import org.tensorflow.proto.framework.GraphDef;
+import org.tensorflow.util.TensorList;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 
@@ -42,7 +43,7 @@ public void graphDefRoundTrip() {
       Ops tf = Ops.create(g);
       tf.withName("Y").linalg.matMul(
           tf.withName("A").constant(new int[2][2]),
-          tf.withName("X").placeholder(TInt32.DTYPE),
+          tf.withName("X").placeholder(TInt32.class),
           MatMul.transposeA(true).transposeB(false)
       );
       graphDef = g.toGraphDef();
@@ -140,8 +141,8 @@ public void addGradientsToGraph() {
         Session s = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<TFloat32> x1 = tf.placeholder(TFloat32.DTYPE).output();
-      Output<TFloat32> x2 = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x1 = tf.placeholder(TFloat32.class).output();
+      Output<TFloat32> x2 = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y0 = tf.math.square(x1).y();
       Output<TFloat32> y1 = tf.math.square(y0).y();
       Output<TFloat32> y2 = tf.math.addN(Arrays.asList(y0, x2)).sum();
@@ -149,29 +150,28 @@ public void addGradientsToGraph() {
       Output<?>[] grads0 = g.addGradients(y1, toArray(x1));
       assertNotNull(grads0);
       assertEquals(1, grads0.length);
-      assertEquals(TFloat32.DTYPE, grads0[0].dataType());
+      assertEquals(TFloat32.class, grads0[0].type());
 
       Output<?>[] grads1 = g.addGradients(y2, toArray(x1, x2));
       assertNotNull(grads1);
       assertEquals(2, grads1.length);
-      assertEquals(TFloat32.DTYPE, grads1[0].dataType());
-      assertEquals(TFloat32.DTYPE, grads1[1].dataType());
+      assertEquals(TFloat32.class, grads1[0].type());
+      assertEquals(TFloat32.class, grads1[1].type());
       
-      try (Tensor<TFloat32> c1 = TFloat32.scalarOf(3.0f);
-          Tensor<TFloat32> c2 = TFloat32.scalarOf(2.0f);
-          AutoCloseableList<Tensor<?>> outputs = new AutoCloseableList<>(
-              s.runner()
+      try (TFloat32 c1 = TFloat32.scalarOf(3.0f);
+          TFloat32 c2 = TFloat32.scalarOf(2.0f);
+          TensorList outputs = s.runner()
                   .feed(x1, c1)
                   .feed(x2, c2)
                   .fetch(grads0[0])
                   .fetch(grads1[0])
                   .fetch(grads1[1])
-                  .run())) {
+                  .run()) {
      
         assertEquals(3, outputs.size());
-        assertEquals(108.0f, outputs.get(0).expect(TFloat32.DTYPE).data().getFloat(), 0.0f);
-        assertEquals(6.0f, outputs.get(1).expect(TFloat32.DTYPE).data().getFloat(), 0.0f);
-        assertEquals(1.0f, outputs.get(2).expect(TFloat32.DTYPE).data().getFloat(), 0.0f);
+        assertEquals(108.0f, ((TFloat32)outputs.get(0)).getFloat(), 0.0f);
+        assertEquals(6.0f, ((TFloat32)outputs.get(1)).getFloat(), 0.0f);
+        assertEquals(1.0f, ((TFloat32)outputs.get(2)).getFloat(), 0.0f);
       }
     }
   }
@@ -182,23 +182,22 @@ public void addGradientSumsToGraph() {
         Session s = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<TFloat32> x = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y0 = tf.math.square(x).y();
       Output<TFloat32> y1 = tf.math.square(y0).y();
 
       Output<?>[] grad = g.addGradients(null, toArray(y0, y1), toArray(x), null);
       assertNotNull(grad);
       assertEquals(1, grad.length);
-      assertEquals(TFloat32.DTYPE, grad[0].dataType());
+      assertEquals(TFloat32.class, grad[0].type());
 
-      try (Tensor<TFloat32> c = TFloat32.scalarOf(3.0f);
-          Tensor<TFloat32> output = s.runner()
+      try (TFloat32 c = TFloat32.scalarOf(3.0f);
+          TFloat32 output = s.runner()
               .feed(x, c)
               .fetch(grad[0])
               .run()
-              .get(0)
-              .expect(TFloat32.DTYPE)) {
-        assertEquals(114.0f, output.data().getFloat(), 0.0f);
+              .single()) {
+        assertEquals(114.0f, output.getFloat(), 0.0f);
       }
     }
   }
@@ -209,28 +208,27 @@ public void addGradientsWithInitialValuesToGraph() {
         Session s = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<TFloat32> x = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y0 = tf.math.square(x).y();
       Output<TFloat32> y1 = tf.math.square(y0).y();
       
       Output<?>[] grad0 = g.addGradients(y1, toArray(y0));
       assertNotNull(grad0);
       assertEquals(1, grad0.length);
-      assertEquals(TFloat32.DTYPE, grad0[0].dataType());
+      assertEquals(TFloat32.class, grad0[0].type());
 
       Output<?>[] grad1 = g.addGradients(null, toArray(y0), toArray(x), toArray(grad0[0]));
       assertNotNull(grad1);
       assertEquals(1, grad1.length);
-      assertEquals(TFloat32.DTYPE, grad1[0].dataType());
+      assertEquals(TFloat32.class, grad1[0].type());
 
-      try (Tensor<TFloat32> c = TFloat32.scalarOf(3.0f);
-          Tensor<TFloat32> output = s.runner()
+      try (TFloat32 c = TFloat32.scalarOf(3.0f);
+          TFloat32 output = s.runner()
               .feed(x, c)
               .fetch(grad1[0])
               .run()
-              .get(0)
-              .expect(TFloat32.DTYPE)) {
-        assertEquals(108.0f, output.data().getFloat(), 0.0f);
+              .single()) {
+        assertEquals(108.0f, output.getFloat(), 0.0f);
       }
     }
   }
@@ -240,7 +238,7 @@ public void validateGradientsNames() {
     try (Graph g = new Graph()) {
       Ops tf = Ops.create(g);
 
-      Output<TFloat32> x = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y0 = tf.math.square(x).y();
 
       Output<?>[] grad0 = g.addGradients(null, toArray(y0), toArray(x), null);
@@ -269,7 +267,7 @@ public void buildWhileLoopSingleInput() {
         Session s = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<?> input = tf.placeholder(TInt32.DTYPE).output();
+      Output<?> input = tf.placeholder(TInt32.class).output();
 
       @SuppressWarnings("unchecked")
       Output<?>[] loopOutputs = g.whileLoop(
@@ -284,14 +282,13 @@ public void buildWhileLoopSingleInput() {
           },
           "test_loop");
 
-      try (Tensor<TInt32> c = TInt32.scalarOf(2);
-          Tensor<TInt32> output = s.runner()
+      try (TInt32 c = TInt32.scalarOf(2);
+          TInt32 output = s.runner()
               .feed(input, c)
               .fetch(loopOutputs[0])
               .run()
-              .get(0)
-              .expect(TInt32.DTYPE)) {
-        assertEquals(16, output.data().getInt()); // ((2^2)^2)
+              .single()) {
+        assertEquals(16, output.getInt()); // ((2^2)^2)
       }
     }
   }
@@ -302,8 +299,8 @@ public void buildWhileLoopMultipleInputs() {
         Session s = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<?> input1 = tf.placeholder(TInt32.DTYPE).output();
-      Output<?> input2 = tf.placeholder(TInt32.DTYPE).output();
+      Output<?> input1 = tf.placeholder(TInt32.class).output();
+      Output<?> input2 = tf.placeholder(TInt32.class).output();
       Output<?>[] inputs = toArray(input1, input2);
 
       @SuppressWarnings("unchecked")
@@ -320,19 +317,17 @@ public void buildWhileLoopMultipleInputs() {
           },
           "test_loop");
 
-      try (Tensor<TInt32> c1 = TInt32.scalarOf(2);
-          Tensor<TInt32> c2 = TInt32.scalarOf(5);
-          AutoCloseableList<Tensor<?>> outputs =
-              new AutoCloseableList<>(
-                  s.runner()
-                      .feed(input1, c1)
-                      .feed(input2, c2)
-                      .fetch(loopOutputs[0])
-                      .fetch(loopOutputs[1])
-                      .run())) {
+      try (TInt32 c1 = TInt32.scalarOf(2);
+          TInt32 c2 = TInt32.scalarOf(5);
+          TensorList outputs = s.runner()
+                  .feed(input1, c1)
+                  .feed(input2, c2)
+                  .fetch(loopOutputs[0])
+                  .fetch(loopOutputs[1])
+                  .run()) {
         assertEquals(2, outputs.size());
-        assertEquals(16, outputs.get(0).expect(TInt32.DTYPE).data().getInt()); // ((2^2)^2)
-        assertEquals(625, outputs.get(1).expect(TInt32.DTYPE).data().getInt()); // ((5^2)^2)
+        assertEquals(16, ((TInt32)outputs.get(0)).getInt()); // ((2^2)^2)
+        assertEquals(625, ((TInt32)outputs.get(1)).getInt()); // ((5^2)^2)
       }
     }
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SavedModelBundleTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SavedModelBundleTest.java
index d807d13de00..67a0f9d105b 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SavedModelBundleTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SavedModelBundleTest.java
@@ -44,6 +44,7 @@
 import org.tensorflow.proto.framework.RunOptions;
 import org.tensorflow.proto.framework.SignatureDef;
 import org.tensorflow.proto.framework.TensorInfo;
+import org.tensorflow.util.TensorMap;
 import org.tensorflow.types.TFloat32;
 
 /** Unit tests for {@link org.tensorflow.SavedModelBundle}. */
@@ -107,9 +108,9 @@ public void exportFunctionWithVariables() throws IOException {
       f.session().run(Init.DEFAULT_NAME);
 
       // Call the graph and remember the result of computation for later
-      try (Tensor<TFloat32> xTensor = TFloat32.tensorOf(xValue);
-          Tensor<TFloat32> zTensor = f.call(xTensor).expect(TFloat32.DTYPE)) {
-        reducedSum = zTensor.data().getFloat();
+      try (TFloat32 xTensor = TFloat32.tensorOf(xValue);
+          TFloat32 zTensor = f.call(xTensor)) {
+        reducedSum = zTensor.getFloat();
       }
       // Save/export the model (which is a single function in this case)
       f.save(testFolder.toString());
@@ -153,15 +154,15 @@ public void exportFunctionWithVariables() throws IOException {
       assertNotNull(outputInfo);
       assertEquals(0, outputInfo.getTensorShape().getDimCount());
 
-      try (Tensor<TFloat32> xTensor = TFloat32.tensorOf(xValue)) {
+      try (TFloat32 xTensor = TFloat32.tensorOf(xValue)) {
         // Call the saved model function and make sure it returns the same result as before
-        try (Tensor<TFloat32> zTensor = function.call(xTensor).expect(TFloat32.DTYPE)) {
-          assertEquals(reducedSum, zTensor.data().getFloat(), EPSILON);
+        try (TFloat32 zTensor = function.call(xTensor)) {
+          assertEquals(reducedSum, zTensor.getFloat(), EPSILON);
         }
         // Now call the same function directly from the model
-        try (Tensor<TFloat32> zTensor =
-            savedModel.call(Collections.singletonMap("input", xTensor)).get("reducedSum").expect(TFloat32.DTYPE)) {
-          assertEquals(reducedSum, zTensor.data().getFloat(), EPSILON);
+        try (TFloat32 zTensor =
+            savedModel.call(Collections.singletonMap("input", xTensor)).get("reducedSum")) {
+          assertEquals(reducedSum, zTensor.getFloat(), EPSILON);
         }
       }
     }
@@ -179,9 +180,9 @@ public void exportMultipleFunctions() throws IOException {
           ConcreteFunction f1 = ConcreteFunction.create(f1Signature, s);
           ConcreteFunction f2 = ConcreteFunction.create(f2Signature, s)) {
         f1.session().run(Init.DEFAULT_NAME);
-        try (Tensor<TFloat32> x = TFloat32.tensorOf(StdArrays.ndCopyOf(new float[]{2, 2}));
-            Tensor<TFloat32> t = f1.call(x).expect(TFloat32.DTYPE)) {
-          reducedSum = t.data().getFloat();
+        try (TFloat32 x = TFloat32.tensorOf(StdArrays.ndCopyOf(new float[]{2, 2}));
+            TFloat32 t = f1.call(x)) {
+          reducedSum = t.getFloat();
         }
         SavedModelBundle.exporter(testFolder.toString())
             .withFunction(f1)
@@ -193,15 +194,15 @@ public void exportMultipleFunctions() throws IOException {
       assertEquals(2, model.signatures().size());
       ConcreteFunction f1 = model.function(Signature.DEFAULT_KEY);
       assertNotNull(f1);
-      try (Tensor<TFloat32> x = TFloat32.tensorOf(StdArrays.ndCopyOf(new float[]{2, 2}));
-          Tensor<TFloat32> t = f1.call(x).expect(TFloat32.DTYPE)) {
-        assertEquals(reducedSum, t.data().getFloat(), EPSILON);
+      try (TFloat32 x = TFloat32.tensorOf(StdArrays.ndCopyOf(new float[]{2, 2}));
+          TFloat32 t = f1.call(x)) {
+        assertEquals(reducedSum, t.getFloat(), EPSILON);
       }
       ConcreteFunction f2 = model.function("identity");
       assertNotNull(f2);
-      try (Tensor<TFloat32> x = TFloat32.scalarOf(10.0f);
-          Tensor<TFloat32> t = f2.call(x).expect(TFloat32.DTYPE)) {
-        assertEquals(10.0f, t.data().getFloat(), 0.0f);
+      try (TFloat32 x = TFloat32.scalarOf(10.0f);
+          TFloat32 t = f2.call(x)) {
+        assertEquals(10.0f, t.getFloat(), 0.0f);
       }
       try {
         model.function("NoSuchFunction");
@@ -290,31 +291,29 @@ public void pythonTfFunction() {
        *   Signature name used for saving 'add', argument names 'a' and 'b'
        */
       ConcreteFunction add = bundle.function("add");
-      Map<String, Tensor<?>> args = new HashMap();
-      try (Tensor<TFloat32> a = TFloat32.scalarOf(10.0f);
-           Tensor<TFloat32> b = TFloat32.scalarOf(15.5f)) {
-        args.put("a", a);
-        args.put("b", b);
-        Map<String, Tensor<?>> result = add.call(args);
+      try (TensorMap args = new TensorMap()) {
+        args.put("a", TFloat32.scalarOf(10.0f));
+        args.put("b", TFloat32.scalarOf(15.5f));
+        TensorMap result = add.call(args);
         assertEquals(result.size(), 1);
-        try (Tensor<TFloat32> c = result.values().iterator().next().expect(TFloat32.DTYPE)) {
-          assertEquals(25.5f, c.data().getFloat());
+        try (TFloat32 c = result.single()) {
+          assertEquals(25.5f, c.getFloat());
         }
       }
     }
   }
 
   private static Signature buildGraphWithVariables(Ops tf, Shape xShape) {
-    Placeholder<TFloat32> x = tf.placeholder(TFloat32.DTYPE, Placeholder.shape(xShape));
+    Placeholder<TFloat32> x = tf.placeholder(TFloat32.class, Placeholder.shape(xShape));
     Variable<TFloat32> y = tf
-        .variable(tf.random.randomUniform(tf.constant(xShape), TFloat32.DTYPE));
+        .variable(tf.random.randomUniform(tf.constant(xShape), TFloat32.class));
     ReduceSum<TFloat32> z = tf.reduceSum(tf.math.add(x, y), tf.array(0, 1));
     Init init = tf.init();
     return Signature.builder().input("input", x).output("reducedSum", z).build();
   }
 
   private static Signature buildIdentityGraph(Ops tf, String signatureKey) {
-    Placeholder<TFloat32> x = tf.placeholder(TFloat32.DTYPE, Placeholder.shape(Shape.scalar()));
+    Placeholder<TFloat32> x = tf.placeholder(TFloat32.class, Placeholder.shape(Shape.scalar()));
     Identity<TFloat32> xprime = tf.identity(x);
     return Signature.builder().key(signatureKey).input("x", x).output("x", xprime).build();
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SessionTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SessionTest.java
index fa41af32a29..cb179ac34ff 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SessionTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/SessionTest.java
@@ -23,7 +23,6 @@
 import java.io.IOException;
 import java.nio.file.Files;
 import java.nio.file.Path;
-import java.nio.file.Paths;
 import org.junit.jupiter.api.Test;
 import org.tensorflow.op.Ops;
 import org.tensorflow.op.core.Init;
@@ -36,6 +35,7 @@
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.StdArrays;
+import org.tensorflow.util.TensorList;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
 
@@ -48,11 +48,10 @@ public void runUsingOperationNames() {
         Session s = new Session(g)) {
       Ops tf = Ops.create(g);
       transpose_A_times_X(tf, new int[][] {{2}, {3}});
-      try (Tensor<TInt32> x = TInt32.tensorOf(StdArrays.ndCopyOf(new int[][] {{5}, {7}}));
-          AutoCloseableList<Tensor<?>> outputs =
-              new AutoCloseableList<>(s.runner().feed("X", x).fetch("Y").run())) {
+      try (TInt32 x = TInt32.tensorOf(StdArrays.ndCopyOf(new int[][] {{5}, {7}}));
+          TensorList outputs = s.runner().feed("X", x).fetch("Y").run()) {
         assertEquals(1, outputs.size());
-        assertEquals(31, outputs.get(0).expect(TInt32.DTYPE).data().getInt(0, 0));
+        assertEquals(31, ((TInt32)outputs.single()).getInt(0, 0));
       }
     }
   }
@@ -65,11 +64,10 @@ public void runUsingOperationHandles() {
       transpose_A_times_X(tf, new int[][] {{2}, {3}});
       Output<TInt32> feed = g.operation("X").output(0);
       Output<TInt32> fetch = g.operation("Y").output(0);
-      try (Tensor<TInt32> x = TInt32.tensorOf(StdArrays.ndCopyOf(new int[][] {{5}, {7}}));
-          AutoCloseableList<Tensor<?>> outputs =
-              new AutoCloseableList<>(s.runner().feed(feed, x).fetch(fetch).run())) {
+      try (TInt32 x = TInt32.tensorOf(StdArrays.ndCopyOf(new int[][] {{5}, {7}}));
+          TensorList outputs = s.runner().feed(feed, x).fetch(fetch).run()) {
         assertEquals(1, outputs.size());
-        assertEquals(31, outputs.get(0).expect(TInt32.DTYPE).data().getInt(0, 0));
+        assertEquals(31, ((TInt32)outputs.single()).getInt(0, 0));
       }
     }
   }
@@ -83,22 +81,20 @@ public void runUsingColonSeparatedNames() {
       tf.math.add(split.output().get(0), split.output().get(1));
 
       // Fetch using colon separated names.
-      try (Tensor<TInt32> fetched =
-          s.runner().fetch("Split:1").run().get(0).expect(TInt32.DTYPE)) {
-        assertEquals(3, fetched.data().getInt(0));
-        assertEquals(4, fetched.data().getInt(1));
+      try (TInt32 fetched = s.runner().fetch("Split:1").run().get(0)) {
+        assertEquals(3, fetched.getInt(0));
+        assertEquals(4, fetched.getInt(1));
       }
       // Feed using colon separated names.
-      try (Tensor<TInt32> fed = TInt32.vectorOf(4, 3, 2, 1);
-          Tensor<TInt32> fetched =
+      try (TInt32 fed = TInt32.vectorOf(4, 3, 2, 1);
+          TInt32 fetched =
               s.runner()
                   .feed("Split:0", fed)
                   .feed("Split:1", fed)
                   .fetch("Add")
                   .run()
-                  .get(0)
-                  .expect(TInt32.DTYPE)) {
-        assertEquals(NdArrays.vectorOf(8, 6, 4, 2), fetched.data());
+                  .single()) {
+        assertEquals(NdArrays.vectorOf(8, 6, 4, 2), fetched);
       }
     }
   }
@@ -109,7 +105,7 @@ public void runWithMetadata() {
         Session s = new Session(g)) {
       Ops tf = Ops.create(g);
       transpose_A_times_X(tf, new int[][] {{2}, {3}});
-      try (Tensor<TInt32> x = TInt32.tensorOf(StdArrays.ndCopyOf(new int[][] {{5}, {7}}))) {
+      try (TInt32 x = TInt32.tensorOf(StdArrays.ndCopyOf(new int[][] {{5}, {7}}))) {
         Session.Run result =
             s.runner()
                 .feed("X", x)
@@ -117,9 +113,9 @@ public void runWithMetadata() {
                 .setOptions(fullTraceRunOptions())
                 .runAndFetchMetadata();
         // Sanity check on outputs.
-        AutoCloseableList<Tensor<?>> outputs = new AutoCloseableList<>(result.outputs);
+        TensorList outputs = result.outputs;
         assertEquals(1, outputs.size());
-        assertEquals(31, outputs.get(0).expect(TInt32.DTYPE).data().getInt(0, 0));
+        assertEquals(31, ((TInt32)outputs.single()).getInt(0, 0));
         // Sanity check on metadata
         assertNotNull(result.metadata);
         assertTrue(result.metadata.hasStepStats(), result.metadata.toString());
@@ -135,11 +131,10 @@ public void runMultipleOutputs() {
       Ops tf = Ops.create(g);
       tf.withName("c1").constant(2718);
       tf.withName("c2").constant(31415);
-      AutoCloseableList<Tensor<?>> outputs =
-          new AutoCloseableList<>(s.runner().fetch("c2").fetch("c1").run());
+      TensorList outputs = s.runner().fetch("c2").fetch("c1").run();
       assertEquals(2, outputs.size());
-      assertEquals(31415, outputs.get(0).expect(TInt32.DTYPE).data().getInt());
-      assertEquals(2718, outputs.get(1).expect(TInt32.DTYPE).data().getInt());
+      assertEquals(31415, ((TInt32)outputs.get(0)).getInt());
+      assertEquals(2718, ((TInt32)outputs.get(1)).getInt());
       outputs.close();
     }
   }
@@ -169,7 +164,7 @@ public void runInit() {
     try (Graph g = new Graph()) {
       Ops tf = Ops.create(g);
 
-      Variable<TInt32> var1 = tf.variable(Shape.scalar(), TInt32.DTYPE);
+      Variable<TInt32> var1 = tf.variable(Shape.scalar(), TInt32.class);
       tf.initAdd(tf.assign(var1, tf.constant(10)));
       Variable<TInt32> var2 = tf.variable(tf.constant(20));
       Add<TInt32> add = tf.math.add(var1, var2);
@@ -177,8 +172,8 @@ public void runInit() {
       try (Session s = new Session(g)) {
         s.run(tf.init());
 
-        try (Tensor<TInt32> t = s.runner().fetch(add).run().get(0).expect(TInt32.DTYPE)) {
-          assertEquals(30, t.data().getInt());
+        try (TInt32 t = s.runner().fetch(add).run().single()) {
+          assertEquals(30, t.getInt());
         }
       }
     }
@@ -189,7 +184,7 @@ public void runInitByName() {
     try (Graph g = new Graph()) {
       Ops tf = Ops.create(g);
 
-      Variable<TInt32> var1 = tf.variable(Shape.scalar(), TInt32.DTYPE);
+      Variable<TInt32> var1 = tf.variable(Shape.scalar(), TInt32.class);
       tf.initAdd(tf.assign(var1, tf.constant(10)));
       Variable<TInt32> var2 = tf.variable(tf.constant(20));
       Add<TInt32> add = tf.math.add(var1, var2);
@@ -198,8 +193,8 @@ public void runInitByName() {
       try (Session s = new Session(g)) {
         s.run("init_test");
 
-        try (Tensor<TInt32> t = s.runner().fetch(add).run().get(0).expect(TInt32.DTYPE)) {
-          assertEquals(30, t.data().getInt());
+        try (TInt32 t = s.runner().fetch(add).run().single()) {
+          assertEquals(30, t.getInt());
         }
         try {
           s.run("wrong_name");
@@ -216,8 +211,8 @@ public void save() throws IOException  {
     Path testFolder = Files.createTempDirectory("tf-session-save-test");
     try (Graph g = new Graph()) {
       Ops tf = Ops.create(g);
-      Variable<TFloat32> x = tf.variable(tf.random.randomUniform(tf.constant(Shape.of(3, 3L)), TFloat32.DTYPE));
-      Variable<TFloat32> y = tf.variable(tf.random.randomUniform(tf.constant(Shape.of(3, 3L)), TFloat32.DTYPE));
+      Variable<TFloat32> x = tf.variable(tf.random.randomUniform(tf.constant(Shape.of(3, 3L)), TFloat32.class));
+      Variable<TFloat32> y = tf.variable(tf.random.randomUniform(tf.constant(Shape.of(3, 3L)), TFloat32.class));
       Init init = tf.init();
 
       try (Session s = new Session(g)) {
@@ -245,7 +240,7 @@ private static ConfigProto singleThreadConfigProto() {
   private static void transpose_A_times_X(Ops tf, int[][] a) {
     tf.withName("Y").linalg.matMul(
         tf.withName("A").constant(a),
-        tf.withName("X").placeholder(TInt32.DTYPE),
+        tf.withName("X").placeholder(TInt32.class),
         MatMul.transposeA(true).transposeB(false)
     );
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/TensorTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/TensorTest.java
index 01ef11efedd..601cf2e6de5 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/TensorTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/TensorTest.java
@@ -30,9 +30,6 @@
 import java.nio.IntBuffer;
 import java.nio.LongBuffer;
 import org.junit.jupiter.api.Test;
-import org.tensorflow.op.Ops;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.buffer.DataBuffers;
 import org.tensorflow.ndarray.BooleanNdArray;
 import org.tensorflow.ndarray.DoubleNdArray;
 import org.tensorflow.ndarray.FloatNdArray;
@@ -40,7 +37,10 @@
 import org.tensorflow.ndarray.LongNdArray;
 import org.tensorflow.ndarray.NdArray;
 import org.tensorflow.ndarray.NdArrays;
+import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.StdArrays;
+import org.tensorflow.ndarray.buffer.DataBuffers;
+import org.tensorflow.op.Ops;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TFloat64;
@@ -48,8 +48,10 @@
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TString;
 import org.tensorflow.types.TUint8;
+import org.tensorflow.types.Type;
+import org.tensorflow.types.TypeRegistry;
 
-/** Unit tests for {@link org.tensorflow.Tensor}. */
+/** Unit tests for {@link Tensor}. */
 public class TensorTest {
   private static final double EPSILON = 1e-7;
   private static final float EPSILON_F = 1e-7f;
@@ -64,22 +66,22 @@ public void createWithRawData() {
     String strings = "test";
     Shape strings_shape = Shape.scalar();
     byte[] strings_; // raw TF_STRING
-    try (Tensor<TString> t = TString.tensorOf(NdArrays.scalarOfObject(strings))) {
+    try (TString t = TString.tensorOf(NdArrays.scalarOfObject(strings))) {
       strings_ = new byte[(int)t.numBytes()];
       t.rawData().read(strings_);
     }
 
     // validate creating a tensor using a raw data byte buffers
     {
-      try (Tensor<TBool> t = Tensor.of(TBool.DTYPE, bools_shape, DataBuffers.of(bools_))) {
+      try (TBool t = Tensors.of(TBool.class, bools_shape, DataBuffers.of(bools_))) {
         boolean[] actual = new boolean[bools_.length];
-        t.data().read(DataBuffers.of(actual));
+        t.read(DataBuffers.of(actual));
         assertArrayEquals(bools, actual);
       }
 
       // note: the buffer is expected to contain raw TF_STRING (as per C API)
-      try (Tensor<TString> t = Tensor.of(TString.DTYPE, strings_shape, DataBuffers.of(strings_))) {
-        assertEquals(strings, t.data().getObject());
+      try (TString t = Tensors.of(TString.class, strings_shape, DataBuffers.of(strings_))) {
+        assertEquals(strings, t.getObject());
       }
     }
 
@@ -87,15 +89,15 @@ public void createWithRawData() {
     {
       DoubleBuffer buf = ByteBuffer.allocateDirect(8 * doubles.length).order(ByteOrder.nativeOrder())
           .asDoubleBuffer().put(doubles);
-      try (Tensor<TFloat64> t = TFloat64.tensorOf(doubles_shape, d -> d.write(DataBuffers.of(buf)))) {
+      try (TFloat64 t = TFloat64.tensorOf(doubles_shape, d -> d.write(DataBuffers.of(buf)))) {
         double[] actual = new double[doubles.length];
-        t.data().read(DataBuffers.of(actual));
+        t.read(DataBuffers.of(actual));
         assertArrayEquals(doubles, actual, EPSILON);
       }
     }
 
     // validate shape checking
-    try (Tensor<TBool> t = Tensor.of(TBool.DTYPE, Shape.of(bools_.length * 2), DataBuffers.of(bools_))) {
+    try (TBool t = Tensors.of(TBool.class, Shape.of(bools_.length * 2), DataBuffers.of(bools_))) {
       fail("should have failed on incompatible buffer");
     } catch (IllegalArgumentException e) {
       // expected
@@ -111,9 +113,9 @@ public void createFromBufferWithNativeByteOrder() {
             .asDoubleBuffer()
             .put(doubles);
     flipBuffer(buf);
-    try (Tensor<TFloat64> t = TFloat64.tensorOf(Shape.of(4), DataBuffers.of(buf))) {
+    try (TFloat64 t = TFloat64.tensorOf(Shape.of(4), DataBuffers.of(buf))) {
       double[] actual = new double[doubles.length];
-      t.data().read(DataBuffers.of(actual));
+      t.read(DataBuffers.of(actual));
       assertArrayEquals(doubles, actual, EPSILON);
     }
   }
@@ -130,9 +132,9 @@ public void createFromBufferWithNonNativeByteOrder() {
             .asDoubleBuffer()
             .put(doubles);
     flipBuffer(buf);
-    try (Tensor<TFloat64> t = TFloat64.tensorOf(Shape.of(4), DataBuffers.of(buf))) {
+    try (TFloat64 t = TFloat64.tensorOf(Shape.of(4), DataBuffers.of(buf))) {
       double[] actual = new double[doubles.length];
-      t.data().read(DataBuffers.of(actual));
+      t.read(DataBuffers.of(actual));
       assertArrayEquals(doubles, actual, EPSILON);
     }
   }
@@ -147,24 +149,24 @@ public void createWithTypedBuffer() {
     // validate creating a tensor using a typed buffer
     {
       Shape shape = Shape.of(4);
-      try (Tensor<TFloat64> t = TFloat64.tensorOf(shape, DataBuffers.of(doubles))) {
+      try (TFloat64 t = TFloat64.tensorOf(shape, DataBuffers.of(doubles))) {
         DoubleBuffer actual = DoubleBuffer.allocate(doubles.capacity());
-        t.data().read(DataBuffers.of(actual));
+        t.read(DataBuffers.of(actual));
         assertEquals(doubles, actual);
       }
-      try (Tensor<TFloat32> t = TFloat32.tensorOf(shape, DataBuffers.of(floats))) {
+      try (TFloat32 t = TFloat32.tensorOf(shape, DataBuffers.of(floats))) {
         FloatBuffer actual = FloatBuffer.allocate(floats.capacity());
-        t.data().read(DataBuffers.of(actual));
+        t.read(DataBuffers.of(actual));
         assertEquals(floats, actual);
       }
-      try (Tensor<TInt32> t = TInt32.tensorOf(shape, DataBuffers.of(ints))) {
+      try (TInt32 t = TInt32.tensorOf(shape, DataBuffers.of(ints))) {
         IntBuffer actual = IntBuffer.allocate(ints.capacity());
-        t.data().read(DataBuffers.of(actual));
+        t.read(DataBuffers.of(actual));
         assertEquals(ints, actual);
       }
-      try (Tensor<TInt64> t = TInt64.tensorOf(shape, DataBuffers.of(longs))) {
+      try (TInt64 t = TInt64.tensorOf(shape, DataBuffers.of(longs))) {
         LongBuffer actual = LongBuffer.allocate(longs.capacity());
-        t.data().read(DataBuffers.of(actual));
+        t.read(DataBuffers.of(actual));
         assertEquals(longs, actual);
       }
     }
@@ -172,22 +174,22 @@ public void createWithTypedBuffer() {
     // validate shape-checking
     {
       Shape shape = Shape.of(5);
-      try (Tensor<TFloat64> t = TFloat64.tensorOf(shape, DataBuffers.of(doubles))) {
+      try (TFloat64 t = TFloat64.tensorOf(shape, DataBuffers.of(doubles))) {
         fail("should have failed on incompatible buffer");
       } catch (BufferUnderflowException e) {
         // expected
       }
-      try (Tensor<TFloat32> t = TFloat32.tensorOf(shape, DataBuffers.of(floats))) {
+      try (TFloat32 t = TFloat32.tensorOf(shape, DataBuffers.of(floats))) {
         fail("should have failed on incompatible buffer");
       } catch (BufferUnderflowException e) {
         // expected
       }
-      try (Tensor<TInt32> t = TInt32.tensorOf(shape, DataBuffers.of(ints))) {
+      try (TInt32 t = TInt32.tensorOf(shape, DataBuffers.of(ints))) {
         fail("should have failed on incompatible buffer");
       } catch (BufferUnderflowException e) {
         // expected
       }
-      try (Tensor<TInt64> t = TInt64.tensorOf(shape, DataBuffers.of(longs))) {
+      try (TInt64 t = TInt64.tensorOf(shape, DataBuffers.of(longs))) {
         fail("should have failed on incompatible buffer");
       } catch (BufferUnderflowException e) {
         // expected
@@ -203,11 +205,11 @@ public void readFromRawData() {
     long[] longs = {1L, 2L, 3L};
     boolean[] bools = {true, false, true};
 
-    try (Tensor<TInt32> tints = TInt32.vectorOf(ints);
-        Tensor<TFloat32> tfloats = TFloat32.vectorOf(floats);
-        Tensor<TFloat64> tdoubles = TFloat64.vectorOf(doubles);
-        Tensor<TInt64> tlongs = TInt64.vectorOf(longs);
-        Tensor<TBool> tbools = TBool.vectorOf(bools)) {
+    try (TInt32 tints = TInt32.vectorOf(ints);
+        TFloat32 tfloats = TFloat32.vectorOf(floats);
+        TFloat64 tdoubles = TFloat64.vectorOf(doubles);
+        TInt64 tlongs = TInt64.vectorOf(longs);
+        TBool tbools = TBool.vectorOf(bools)) {
 
       // validate that any datatype is readable with ByteBuffer (content, position)
       {
@@ -266,79 +268,79 @@ public void readFromRawData() {
 
   @Test
   public void scalars() {
-    try (Tensor<TFloat32> t = TFloat32.scalarOf(2.718f)) {
-      assertEquals(TFloat32.DTYPE, t.dataType());
+    try (TFloat32 t = TFloat32.scalarOf(2.718f)) {
+      assertEquals(TFloat32.class, t.type());
       assertEquals(0, t.shape().numDimensions());
-      assertEquals(2.718f, t.data().getFloat(), EPSILON_F);
+      assertEquals(2.718f, t.getFloat(), EPSILON_F);
     }
 
-    try (Tensor<TFloat64> t = TFloat64.scalarOf(3.1415)) {
-      assertEquals(TFloat64.DTYPE, t.dataType());
+    try (TFloat64 t = TFloat64.scalarOf(3.1415)) {
+      assertEquals(TFloat64.class, t.type());
       assertEquals(0, t.shape().numDimensions());
-      assertEquals(3.1415, t.data().getDouble(), EPSILON);
+      assertEquals(3.1415, t.getDouble(), EPSILON);
     }
 
-    try (Tensor<TInt32> t = TInt32.scalarOf(-33)) {
-      assertEquals(TInt32.DTYPE, t.dataType());
+    try (TInt32 t = TInt32.scalarOf(-33)) {
+      assertEquals(TInt32.class, t.type());
       assertEquals(0, t.shape().numDimensions());
-      assertEquals(-33, t.data().getInt());
+      assertEquals(-33, t.getInt());
     }
 
-    try (Tensor<TInt64> t = TInt64.scalarOf(8589934592L)) {
-      assertEquals(TInt64.DTYPE, t.dataType());
+    try (TInt64 t = TInt64.scalarOf(8589934592L)) {
+      assertEquals(TInt64.class, t.type());
       assertEquals(0, t.shape().numDimensions());
-      assertEquals(8589934592L, t.data().getLong());
+      assertEquals(8589934592L, t.getLong());
     }
 
-    try (Tensor<TBool> t = TBool.scalarOf(true)) {
-      assertEquals(TBool.DTYPE, t.dataType());
+    try (TBool t = TBool.scalarOf(true)) {
+      assertEquals(TBool.class, t.type());
       assertEquals(0, t.shape().numDimensions());
-      assertTrue(t.data().getBoolean());
+      assertTrue(t.getBoolean());
     }
 
-    try (Tensor<TString> t = TString.scalarOf("sombrero")) {
-      assertEquals(TString.DTYPE, t.dataType());
+    try (TString t = TString.scalarOf("sombrero")) {
+      assertEquals(TString.class, t.type());
       assertEquals(0, t.shape().numDimensions());
-      assertEquals("sombrero", t.data().getObject());
+      assertEquals("sombrero", t.getObject());
     }
 
     final byte[] bytes = {1, 2, 3, 4};
-    try (Tensor<TString> t = TString.tensorOfBytes(NdArrays.scalarOfObject(bytes))) {
-      assertEquals(TString.DTYPE, t.dataType());
+    try (TString t = TString.tensorOfBytes(NdArrays.scalarOfObject(bytes))) {
+      assertEquals(TString.class, t.type());
       assertEquals(0, t.shape().numDimensions());
-      assertArrayEquals(bytes, t.data().asBytes().getObject());
+      assertArrayEquals(bytes, t.asBytes().getObject());
     }
   }
 
   @Test
   public void nDimensional() {
     DoubleNdArray vector = StdArrays.ndCopyOf(new double[]{1.414, 2.718, 3.1415});
-    try (Tensor<TFloat64> t = TFloat64.tensorOf(vector)) {
-      assertEquals(TFloat64.DTYPE, t.dataType());
+    try (TFloat64 t = TFloat64.tensorOf(vector)) {
+      assertEquals(TFloat64.class, t.type());
       assertEquals(1, t.shape().numDimensions());
       assertEquals(3, t.shape().size(0));
-      assertEquals(vector, t.data());
+      assertEquals(vector, t);
     }
 
     IntNdArray matrix = StdArrays.ndCopyOf(new int[][]{{1, 2, 3}, {4, 5, 6}});
-    try (Tensor<TInt32> t = TInt32.tensorOf(matrix)) {
-      assertEquals(TInt32.DTYPE, t.dataType());
+    try (TInt32 t = TInt32.tensorOf(matrix)) {
+      assertEquals(TInt32.class, t.type());
       assertEquals(2, t.shape().numDimensions());
       assertEquals(2, t.shape().size(0));
       assertEquals(3, t.shape().size(1));
-      assertEquals(matrix, t.data());
+      assertEquals(matrix, t);
     }
 
     LongNdArray threeD = StdArrays.ndCopyOf(new long[][][]{
       {{1}, {3}, {5}, {7}, {9}}, {{2}, {4}, {6}, {8}, {0}},
     });
-    try (Tensor<TInt64> t = TInt64.tensorOf(threeD)) {
-      assertEquals(TInt64.DTYPE, t.dataType());
+    try (TInt64 t = TInt64.tensorOf(threeD)) {
+      assertEquals(TInt64.class, t.type());
       assertEquals(3, t.shape().numDimensions());
       assertEquals(2, t.shape().size(0));
       assertEquals(5, t.shape().size(1));
       assertEquals(1, t.shape().size(2));
-      assertEquals(threeD, t.data());
+      assertEquals(threeD, t);
     }
 
     BooleanNdArray fourD = StdArrays.ndCopyOf(new boolean[][][][]{
@@ -346,14 +348,14 @@ public void nDimensional() {
       {{{false, false, true, true}, {false, true, false, false}}},
       {{{false, true, false, true}, {false, true, true, false}}},
     });
-    try (Tensor<TBool> t = TBool.tensorOf(fourD)) {
-      assertEquals(TBool.DTYPE, t.dataType());
+    try (TBool t = TBool.tensorOf(fourD)) {
+      assertEquals(TBool.class, t.type());
       assertEquals(4, t.shape().numDimensions());
       assertEquals(3, t.shape().size(0));
       assertEquals(1, t.shape().size(1));
       assertEquals(2, t.shape().size(2));
       assertEquals(4, t.shape().size(3));
-      assertEquals(fourD, t.data());
+      assertEquals(fourD, t);
     }
   }
 
@@ -365,36 +367,36 @@ public void testNDimensionalStringTensor() {
         matrix.setObject(String.format("(%d, %d) = %d", i, j, i << j), i, j);
       }
     }
-    try (Tensor<TString> t = TString.tensorOf(matrix)) {
-      assertEquals(TString.DTYPE, t.dataType());
+    try (TString t = TString.tensorOf(matrix)) {
+      assertEquals(TString.class, t.type());
       assertEquals(2, t.shape().numDimensions());
       assertEquals(4, t.shape().size(0));
       assertEquals(3, t.shape().size(1));
-      assertEquals(matrix, t.data());
+      assertEquals(matrix, t);
     }
 
     NdArray<byte[]> byteMatrix = NdArrays.ofObjects(byte[].class, matrix.shape());
     matrix.scalars().forEachIndexed((i, s) -> byteMatrix.setObject(s.getObject().getBytes(UTF_8), i));
-    try (Tensor<TString> t = TString.tensorOfBytes(byteMatrix)) {
-      assertEquals(TString.DTYPE, t.dataType());
+    try (TString t = TString.tensorOfBytes(byteMatrix)) {
+      assertEquals(TString.class, t.type());
       assertEquals(2, t.shape().numDimensions());
       assertEquals(4, t.shape().size(0));
       assertEquals(3, t.shape().size(1));
-      assertEquals(byteMatrix, t.data().asBytes());
-      assertEquals(matrix, t.data());
+      assertEquals(byteMatrix, t.asBytes());
+      assertEquals(matrix, t);
     }
   }
 
   @Test
   public void testUint8TensorFromArray() {
     byte[] vector = new byte[] {1, 2, 3, 4};
-    try (Tensor<TUint8> t = TUint8.vectorOf(vector)) {
-      assertEquals(TUint8.DTYPE, t.dataType());
+    try (TUint8 t = TUint8.vectorOf(vector)) {
+      assertEquals(TUint8.class, t.type());
       assertEquals(1, t.shape().numDimensions());
       assertEquals(4, t.shape().size(0));
 
       byte[] got = new byte[4];
-      t.data().read(DataBuffers.of(got));
+      t.read(DataBuffers.of(got));
       assertArrayEquals(vector, got);
     }
   }
@@ -402,13 +404,13 @@ public void testUint8TensorFromArray() {
   @Test
   public void testCreateFromArrayOfBoxed() {
     Integer[] vector = new Integer[] {1, 2, 3, 4};
-    try (Tensor<TInt32> t = TInt32.tensorOf(Shape.of(4), d -> d.write(DataBuffers.ofObjects(vector)))) {
-      assertEquals(TInt32.DTYPE, t.dataType());
+    try (TInt32 t = TInt32.tensorOf(Shape.of(4), d -> d.write(DataBuffers.ofObjects(vector)))) {
+      assertEquals(TInt32.class, t.type());
       assertEquals(1, t.shape().numDimensions());
       assertEquals(4, t.shape().size(0));
 
       Integer[] got = new Integer[4];
-      t.data().read(DataBuffers.ofObjects(got));
+      t.read(DataBuffers.ofObjects(got));
       assertArrayEquals(vector, got);
     }
   }
@@ -421,7 +423,7 @@ public void failCreateOnMismatchedDimensions() {
         invalid[x][y] = new int[x + y + 1];
       }
     }
-    try (Tensor<TInt32> t = TInt32.tensorOf(StdArrays.ndCopyOf(invalid))) {
+    try (TInt32 t = TInt32.tensorOf(StdArrays.ndCopyOf(invalid))) {
       fail("Tensor.create() should fail because of differing sizes in the 3rd dimension");
     } catch (IllegalArgumentException e) {
       // The expected exception.
@@ -433,11 +435,11 @@ public void tensorWithZeroDimension() {
     // Note: Historically, TF Java failed on purpose when trying to allocate a tensor with a shape
     // that has one or more dimensions set to 0 elements. But Python API allows it, so we should do
     // the same.
-    try (Tensor<TInt32> t = TInt32.tensorOf(Shape.of(3, 0, 1))) {
+    try (TInt32 t = TInt32.tensorOf(Shape.of(3, 0, 1))) {
       assertEquals(0, t.numBytes());
       assertEquals(0, t.shape().size());
     }
-    try (Tensor<TInt32> t = TInt32.tensorOf(StdArrays.ndCopyOf(new int[3][0][1]))) {
+    try (TInt32 t = TInt32.tensorOf(StdArrays.ndCopyOf(new int[3][0][1]))) {
       assertEquals(0, t.numBytes());
       assertEquals(0, t.shape().size());
     }
@@ -445,53 +447,36 @@ public void tensorWithZeroDimension() {
 
   @Test
   public void allocateTensorWithSize() {
-    try (Tensor<TInt32> t = Tensor.of(TInt32.DTYPE, Shape.of(2, 2, 2), 8 * TInt32.DTYPE.byteSize())) {
+    Type<TInt32> type = TypeRegistry.find(TInt32.class);
+    try (TInt32 t = Tensors.of(TInt32.class, Shape.of(2, 2, 2), 8 * type.byteSize())) {
       // ok
     }
-    try (Tensor<TInt32> t = Tensor.of(TInt32.DTYPE, Shape.of(2, 2, 2), 9 * TInt32.DTYPE.byteSize())) {
+    try (TInt32 t = Tensors.of(TInt32.class, Shape.of(2, 2, 2), 9 * type.byteSize())) {
       // ok (size requested is larger that minimum space required)
     }
     try {
-      Tensor.of(TInt32.DTYPE, Shape.of(2, 2, 2), 8 * TInt32.DTYPE.byteSize() - 1);
+      Tensors.of(TInt32.class, Shape.of(2, 2, 2), 8 * type.byteSize() - 1);
       fail();
     } catch (IllegalArgumentException e) {
       // as expected
     }
   }
 
-  @Test
-  public void useAfterClose() {
-    int n = 4;
-    Tensor<?> t = TInt32.scalarOf(n);
-    t.close();
-    try {
-      t.data();
-    } catch (IllegalStateException e) {
-      // The expected exception.
-    }
-  }
-
   @Test
   public void eagerTensorIsReleasedAfterSessionIsClosed() {
-    Tensor<TInt32> sum;
+    TInt32 sum;
     try (EagerSession session = EagerSession.create()) {
       Ops tf = Ops.create(session);
       sum = tf.math.add(tf.constant(10), tf.constant(20)).asTensor();
-      sum.nativeHandle(); // does not throw
-      assertEquals(30, sum.data().getInt());
+      sum.handle().get(); // does not throw
+      assertEquals(30, sum.getInt());
     }
     try {
-      sum.nativeHandle();
+      sum.handle().get();
       fail("Tensor native handle should have been closed by ending eager session");
     } catch (IllegalStateException e) {
       // as expected
     }
-    try {
-      sum.data().getInt();
-      fail("Tensor data should not be accessible after tensor is closed");
-    } catch (IllegalStateException e) {
-      // as expected
-    }
   }
 
   @Test
@@ -503,12 +488,12 @@ public void fromHandle() {
     // An exception is made for this test, where the pitfalls of this is avoided by not calling
     // close() on both Tensors.
     final FloatNdArray matrix = StdArrays.ndCopyOf(new float[][]{{1, 2, 3}, {4, 5, 6}});
-    try (Tensor<TFloat32> src = TFloat32.tensorOf(matrix)) {
-      Tensor<TFloat32> cpy = Tensor.fromHandle(src.nativeHandle()).expect(TFloat32.DTYPE);
-      assertEquals(src.dataType(), cpy.dataType());
+    try (TFloat32 src = TFloat32.tensorOf(matrix)) {
+      TFloat32 cpy = Tensors.fromHandle(src.handle());
+      assertEquals(src.type(), cpy.type());
       assertEquals(src.shape().numDimensions(), cpy.shape().numDimensions());
       assertEquals(src.shape(), cpy.shape());
-      assertEquals(matrix, cpy.data());
+      assertEquals(matrix, cpy);
     }
   }
 
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/ScopeTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/ScopeTest.java
index bbebfd5f454..e83c956adb7 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/ScopeTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/ScopeTest.java
@@ -23,9 +23,8 @@
 import org.tensorflow.Graph;
 import org.tensorflow.Output;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
-import org.tensorflow.types.TInt32;
 import org.tensorflow.types.family.TType;
+import org.tensorflow.types.TInt32;
 
 /** Unit tests for {@link org.tensorflow.op.Scope}. */
 public class ScopeTest {
@@ -169,11 +168,12 @@ public void composite() {
       // assertNotNull(g.operation("variance/zero"));
 
       // Verify correct results as well.
-      Tensor<TInt32> result =
-          sess.runner().fetch(var1.output()).run().get(0).expect(TInt32.DTYPE);
-      assertEquals(21704, result.data().getInt());
-      result = sess.runner().fetch(var2.output()).run().get(0).expect(TInt32.DTYPE);
-      assertEquals(21704, result.data().getInt());
+      try (TInt32 result = sess.runner().fetch(var1.output()).run().single()) {
+        assertEquals(21704, result.getInt());
+      }
+      try (TInt32 result = sess.runner().fetch(var2.output()).run().single()) {
+        assertEquals(21704, result.getInt());
+      }
     }
   }
 
@@ -189,11 +189,11 @@ static Const<TInt32> create(Scope s, int[] v) {
       return create(s, TInt32.vectorOf(v));
     }
 
-    static <T extends TType> Const<T> create(Scope s, Tensor<T> value) {
+    static <T extends TType> Const<T> create(Scope s, T value) {
       return new Const<>(
           s.env()
               .opBuilder("Const", s.makeOpName("Const"))
-              .setAttr("dtype", value.dataType())
+              .setAttr("dtype", value.type())
               .setAttr("value", value)
               .build()
               .output(0));
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ConstantTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ConstantTest.java
index 266a62bd1ed..0f94a546ca8 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ConstantTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ConstantTest.java
@@ -18,16 +18,10 @@
 import static org.junit.jupiter.api.Assertions.assertEquals;
 
 import java.io.IOException;
-import java.nio.ByteBuffer;
-import java.nio.DoubleBuffer;
-import java.nio.FloatBuffer;
-import java.nio.IntBuffer;
-import java.nio.LongBuffer;
 import org.junit.jupiter.api.Test;
-import org.tensorflow.AutoCloseableList;
 import org.tensorflow.Graph;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
+import org.tensorflow.util.TensorList;
 import org.tensorflow.op.Scope;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.buffer.DataBuffer;
@@ -62,10 +56,9 @@ public void createInts() {
       Scope scope = new Scope(g);
       Constant<TInt32> op1 = Constant.tensorOf(scope, shape, buffer);
       Constant<TInt32> op2 = Constant.tensorOf(scope, array);
-      try (AutoCloseableList<Tensor<?>> t =
-          new AutoCloseableList<>(sess.runner().fetch(op1).fetch(op2).run())) {
-        assertEquals(array, t.get(0).expect(TInt32.DTYPE).data());
-        assertEquals(array, t.get(1).expect(TInt32.DTYPE).data());
+      try (TensorList t = sess.runner().fetch(op1).fetch(op2).run()) {
+        assertEquals(array, t.get(0));
+        assertEquals(array, t.get(1));
       }
     }
   }
@@ -81,10 +74,9 @@ public void createFloats() {
       Scope scope = new Scope(g);
       Constant<TFloat32> op1 = Constant.tensorOf(scope, shape, buffer);
       Constant<TFloat32> op2 = Constant.tensorOf(scope, array);
-      try (AutoCloseableList<Tensor<?>> t =
-          new AutoCloseableList<>(sess.runner().fetch(op1).fetch(op2).run())) {
-        assertEquals(array, t.get(0).expect(TFloat32.DTYPE).data());
-        assertEquals(array, t.get(1).expect(TFloat32.DTYPE).data());
+      try (TensorList t = sess.runner().fetch(op1).fetch(op2).run()) {
+        assertEquals(array, t.get(0));
+        assertEquals(array, t.get(1));
       }
     }
   }
@@ -100,10 +92,9 @@ public void createDoubles() {
       Scope scope = new Scope(g);
       Constant<TFloat64> op1 = Constant.tensorOf(scope, shape, buffer);
       Constant<TFloat64> op2 = Constant.tensorOf(scope, array);
-      try (AutoCloseableList<Tensor<?>> t =
-          new AutoCloseableList<>(sess.runner().fetch(op1).fetch(op2).run())) {
-        assertEquals(array, t.get(0).expect(TFloat64.DTYPE).data());
-        assertEquals(array, t.get(1).expect(TFloat64.DTYPE).data());
+      try (TensorList t = sess.runner().fetch(op1).fetch(op2).run()) {
+        assertEquals(array, t.get(0));
+        assertEquals(array, t.get(1));
       }
     }
   }
@@ -119,10 +110,9 @@ public void createLongs() {
       Scope scope = new Scope(g);
       Constant<TInt64> op1 = Constant.tensorOf(scope, shape, buffer);
       Constant<TInt64> op2 = Constant.tensorOf(scope, array);
-      try (AutoCloseableList<Tensor<?>> t =
-          new AutoCloseableList<>(sess.runner().fetch(op1).fetch(op2).run())) {
-        assertEquals(array, t.get(0).expect(TInt64.DTYPE).data());
-        assertEquals(array, t.get(1).expect(TInt64.DTYPE).data());
+      try (TensorList t = sess.runner().fetch(op1).fetch(op2).run()) {
+        assertEquals(array, t.get(0));
+        assertEquals(array, t.get(1));
       }
     }
   }
@@ -138,10 +128,9 @@ public void createStrings() throws IOException {
       Scope scope = new Scope(g);
       Constant<TString> op1 = Constant.tensorOf(scope, shape, buffer);
       Constant<TString> op2 = Constant.tensorOf(scope, array);
-      try (AutoCloseableList<Tensor<?>> t =
-          new AutoCloseableList<>(sess.runner().fetch(op1).fetch(op2).run())) {
-        assertEquals(array, t.get(0).expect(TString.DTYPE).data());
-        assertEquals(array, t.get(1).expect(TString.DTYPE).data());
+      try (TensorList t = sess.runner().fetch(op1).fetch(op2).run()) {
+        assertEquals(array, t.get(0));
+        assertEquals(array, t.get(1));
       }
     }
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GeneratedOperationsTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GeneratedOperationsTest.java
index a337bd73098..e15583cda8c 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GeneratedOperationsTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GeneratedOperationsTest.java
@@ -22,7 +22,6 @@
 import org.tensorflow.Graph;
 import org.tensorflow.Operand;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.Op;
 import org.tensorflow.op.Ops;
 import org.tensorflow.ndarray.Shape;
@@ -36,8 +35,8 @@ public void tensorInputTensorOutput() {
         Session sess = new Session(g)) {
       Ops ops = Ops.create(g);
       Operand<TInt32> x = ops.math.add(ops.constant(1), ops.constant(2));
-      try (Tensor<TInt32> result = sess.runner().fetch(x).run().get(0).expect(TInt32.DTYPE)) {
-        assertEquals(3, result.data().getInt());
+      try (TInt32 result = sess.runner().fetch(x).run().single()) {
+        assertEquals(3, result.getInt());
       }
     }
   }
@@ -52,8 +51,8 @@ public void testListInputTensorOutput() {
       inputs.add(ops.constant(2));
       inputs.add(ops.constant(3));
       Operand<TInt32> x = ops.math.addN(inputs);
-      try (Tensor<TInt32> result = sess.runner().fetch(x).run().get(0).expect(TInt32.DTYPE)) {
-        assertEquals(6, result.data().getInt());
+      try (TInt32 result = sess.runner().fetch(x).run().single()) {
+        assertEquals(6, result.getInt());
       }
     }
   }
@@ -70,15 +69,15 @@ public void testControlDependencies() {
     try (Graph g = new Graph();
         Session sess = new Session(g)) {
       Ops ops = Ops.create(g);
-      Operand<TInt32> variable = ops.variable(Shape.scalar(), TInt32.DTYPE);
+      Operand<TInt32> variable = ops.variable(Shape.scalar(), TInt32.class);
       Operand<?> initVariable = ops.assign(variable, ops.constant(0));
       ArrayList<Op> controls = new ArrayList<>();
       controls.add(ops.assign(variable, ops.constant(3)));
       Operand<TInt32> x =
           ops.withControlDependencies(controls).math.add(variable, ops.constant(0));
       sess.runner().addTarget(initVariable).run();
-      try (Tensor<TInt32> result = sess.runner().fetch(x).run().get(0).expect(TInt32.DTYPE)) {
-        assertEquals(3, result.data().getInt());
+      try (TInt32 result = sess.runner().fetch(x).run().single()) {
+        assertEquals(3, result.getInt());
       }
     }
   }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GradientsTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GradientsTest.java
index fe1503d415f..df2488662fc 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GradientsTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/GradientsTest.java
@@ -21,12 +21,11 @@
 
 import java.util.Arrays;
 import org.junit.jupiter.api.Test;
-import org.tensorflow.AutoCloseableList;
 import org.tensorflow.Graph;
 import org.tensorflow.Output;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.Ops;
+import org.tensorflow.util.TensorList;
 import org.tensorflow.types.TFloat32;
 
 public class GradientsTest {
@@ -37,7 +36,7 @@ public void createGradients() {
         Session sess = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<TFloat32> x = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y0 = tf.math.square(x).y();
       Output<TFloat32> y1 = tf.math.square(y0).y();
 
@@ -47,13 +46,10 @@ public void createGradients() {
       assertNotNull(grads.dy());
       assertEquals(2, grads.dy().size());
 
-      try (Tensor<TFloat32> c = TFloat32.scalarOf(3.0f);
-          AutoCloseableList<Tensor<?>> outputs =
-              new AutoCloseableList<>(
-                  sess.runner().feed(x, c).fetch(grads.dy(0)).fetch(grads.dy(1)).run())) {
-
-        assertEquals(108.0f, outputs.get(0).expect(TFloat32.DTYPE).data().getFloat(), 0.0f);
-        assertEquals(18.0f, outputs.get(1).expect(TFloat32.DTYPE).data().getFloat(), 0.0f);
+      try (TFloat32 c = TFloat32.scalarOf(3.0f);
+          TensorList outputs = sess.runner().feed(x, c).fetch(grads.dy(0)).fetch(grads.dy(1)).run()) {
+        assertEquals(108.0f, ((TFloat32)outputs.get(0)).getFloat(), 0.0f);
+        assertEquals(18.0f, ((TFloat32)outputs.get(1)).getFloat(), 0.0f);
       }
     }
   }
@@ -64,7 +60,7 @@ public void createGradientsWithSum() {
         Session sess = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<TFloat32> x = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y0 = tf.math.square(x).y();
       Output<TFloat32> y1 = tf.math.square(y0).y();
 
@@ -74,11 +70,9 @@ public void createGradientsWithSum() {
       assertNotNull(grads.dy());
       assertEquals(1, grads.dy().size());
 
-      try (Tensor<TFloat32> c = TFloat32.scalarOf(3.0f);
-          AutoCloseableList<Tensor<?>> outputs =
-              new AutoCloseableList<>(sess.runner().feed(x, c).fetch(grads.dy(0)).run())) {
-
-        assertEquals(114.0f, outputs.get(0).expect(TFloat32.DTYPE).data().getFloat(), 0.0f);
+      try (TFloat32 c = TFloat32.scalarOf(3.0f);
+          TensorList outputs = sess.runner().feed(x, c).fetch(grads.dy(0)).run()) {
+        assertEquals(114.0f, ((TFloat32)outputs.get(0)).getFloat(), 0.0f);
       }
     }
   }
@@ -89,7 +83,7 @@ public void createGradientsWithInitialValues() {
         Session sess = new Session(g)) {
       Ops tf = Ops.create(g);
 
-      Output<TFloat32> x = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y0 = tf.math.square(x).y();
       Output<TFloat32> y1 = tf.math.square(y0).y();
 
@@ -100,12 +94,9 @@ public void createGradientsWithInitialValues() {
       assertNotNull(grads1.dy());
       assertEquals(1, grads1.dy().size());
 
-      try (Tensor<TFloat32> c = TFloat32.scalarOf(3.0f);
-          AutoCloseableList<Tensor<?>> outputs =
-              new AutoCloseableList<>(
-                  sess.runner().feed(x, c).fetch(grads1.dy(0)).run())) {
-
-        assertEquals(108.0f, outputs.get(0).expect(TFloat32.DTYPE).data().getFloat(), 0.0f);
+      try (TFloat32 c = TFloat32.scalarOf(3.0f);
+          TensorList outputs = sess.runner().feed(x, c).fetch(grads1.dy(0)).run()) {
+        assertEquals(108.0f, ((TFloat32)outputs.get(0)).getFloat(), 0.0f);
       }
     }
   }
@@ -115,7 +106,7 @@ public void validateGradientsNames() {
     try (Graph g = new Graph()) {
       Ops tf = Ops.create(g).withSubScope("sub");
 
-      Output<TFloat32> x = tf.placeholder(TFloat32.DTYPE).output();
+      Output<TFloat32> x = tf.placeholder(TFloat32.class).output();
       Output<TFloat32> y = tf.math.square(x).y();
 
       Gradients grad0 = Gradients.create(tf.scope(), y, Arrays.asList(x));
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ShapesTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ShapesTest.java
index d5eb7412ea3..665d599145e 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ShapesTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ShapesTest.java
@@ -20,12 +20,10 @@
 
 import org.junit.jupiter.api.Test;
 
-import org.junit.jupiter.api.TestTemplate;
 import org.tensorflow.Graph;
 import org.tensorflow.EagerSession;
 import org.tensorflow.Operand;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.Scope;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TInt32;
@@ -40,22 +38,19 @@ public void testFlatten_Operand() {
         Session session = new Session(g)) {
       Scope scope = new Scope(g);
       Operand<TFloat32> operand = Constant.arrayOf(scope, new float[] {1, 2, 3, 4, 5, 6, 7, 8});
-      Shape<TInt64> expResult = Shape.create(scope, operand, TInt64.DTYPE);
+      Shape<TInt64> expResult = Shape.create(scope, operand, TInt64.class);
       Operand<TFloat32> reshaped =
           Reshape.create(scope, operand, Constant.vectorOf(scope, new long[] {4, 2, 1}));
       Operand actual = Shapes.flatten(scope, reshaped);
-      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.DTYPE);
+      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.class);
 
       AtomicInteger index = new AtomicInteger();
-      try (Tensor<TInt64> result1 =
-              session.runner().fetch(tfshape.asOutput()).run().get(0).expect(TInt64.DTYPE);
-          Tensor<TInt64> result2 =
-              session.runner().fetch(expResult.asOutput()).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result1 = session.runner().fetch(tfshape.asOutput()).run().single();
+          TInt64 result2 = session.runner().fetch(expResult.asOutput()).run().single()) {
         result1
-            .data()
             .scalars()
             .forEach(
-                s -> assertEquals(result2.data().getLong(index.getAndIncrement()), s.getLong()));
+                s -> assertEquals(result2.getLong(index.getAndIncrement()), s.getLong()));
       }
     }
   }
@@ -66,21 +61,21 @@ public void testFlatten_Shape() {
     try (EagerSession session = EagerSession.create()) {
       Scope scope = new Scope(session);
       Operand operand = Constant.arrayOf(scope, new float[] {1, 2, 3, 4, 5, 6, 7, 8});
-      Shape<TInt64> expShape = Shape.create(scope, operand, TInt64.DTYPE);
+      Shape<TInt64> expShape = Shape.create(scope, operand, TInt64.class);
       Operand actual =
           Reshape.create(scope, operand, Constant.vectorOf(scope, new long[] {4, 2, 1}));
-      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.DTYPE);
-      Operand<TInt64> flattened = Shapes.flatten(scope, tfshape, TInt64.DTYPE);
+      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.class);
+      Operand<TInt64> flattened = Shapes.flatten(scope, tfshape, TInt64.class);
 
       AtomicInteger index = new AtomicInteger();
       flattened
           .asOutput()
-          .data()
+          .asTensor()
           .scalars()
           .forEach(
               s ->
                   assertEquals(
-                      expShape.asOutput().data().getLong(index.getAndIncrement()), s.getLong()));
+                      expShape.asTensor().getLong(index.getAndIncrement()), s.getLong()));
     }
   }
 
@@ -93,13 +88,12 @@ public void testSize_Shape() {
       Operand operand = Constant.arrayOf(scope, new float[] {1, 2, 3, 4, 5, 6, 7, 8});
       Operand actual =
           Reshape.create(scope, operand, Constant.vectorOf(scope, new long[] {4, 2, 1}));
-      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.DTYPE);
-      Operand<TInt64> size = Shapes.size(scope, tfshape, TInt64.DTYPE);
+      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.class);
+      Operand<TInt64> size = Shapes.size(scope, tfshape, TInt64.class);
 
       AtomicInteger index = new AtomicInteger();
-      try (Tensor<TInt64> result1 =
-          session.runner().fetch(size.asOutput()).run().get(0).expect(TInt64.DTYPE)) {
-        result1.data().scalars().forEach(s -> assertEquals(8, s.getLong()));
+      try (TInt64 result1 = session.runner().fetch(size.asOutput()).run().single()) {
+        result1.scalars().forEach(s -> assertEquals(8, s.getLong()));
       }
     }
   }
@@ -116,21 +110,21 @@ public void testSize_Shape_Operand() {
       Shape<TInt32> tfshape = Shape.create(scope, actual);
 
       Operand<TInt32> size = Shapes.size(scope, tfshape, Constant.scalarOf(scope, 0));
-      try (Tensor<TInt32> result =
-          session.runner().fetch(size.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(4, s.getInt()));
+      try (TInt32 result =
+          session.runner().fetch(size.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(4, s.getInt()));
       }
 
       size = Shapes.size(scope, tfshape, Constant.scalarOf(scope, 1));
-      try (Tensor<TInt32> result =
-          session.runner().fetch(size.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(2, s.getInt()));
+      try (TInt32 result =
+          session.runner().fetch(size.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(2, s.getInt()));
       }
 
       size = Shapes.size(scope, tfshape, Constant.scalarOf(scope, 2));
-      try (Tensor<TInt32> result =
-          session.runner().fetch(size.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(1, s.getInt()));
+      try (TInt32 result =
+          session.runner().fetch(size.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(1, s.getInt()));
       }
     }
   }
@@ -146,21 +140,21 @@ public void testSize_Operand_Operand() {
           Reshape.create(scope, operand, Constant.vectorOf(scope, new long[] {4, 2, 1}));
 
       Operand<TInt32> size = Shapes.size(scope, actual, Constant.scalarOf(scope, 0));
-      try (Tensor<TInt32> result =
-          session.runner().fetch(size.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(4, s.getInt()));
+      try (TInt32 result =
+          session.runner().fetch(size.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(4, s.getInt()));
       }
 
       size = Shapes.size(scope, actual, Constant.scalarOf(scope, 1));
-      try (Tensor<TInt32> result =
-          session.runner().fetch(size.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(2, s.getInt()));
+      try (TInt32 result =
+          session.runner().fetch(size.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(2, s.getInt()));
       }
 
       size = Shapes.size(scope, actual, Constant.scalarOf(scope, 2));
-      try (Tensor<TInt32> result =
-          session.runner().fetch(size.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(1, s.getInt()));
+      try (TInt32 result =
+          session.runner().fetch(size.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(1, s.getInt()));
       }
     }
   }
@@ -177,9 +171,9 @@ public void testNumDimensions() {
       Shape<TInt32> tfshape = Shape.create(scope, actual);
 
       Operand<TInt32> nDims = Shapes.numDimensions(scope, tfshape);
-      try (Tensor<TInt32> result =
-          session.runner().fetch(nDims.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(3, s.getInt()));
+      try (TInt32 result =
+          session.runner().fetch(nDims.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(3, s.getInt()));
       }
     }
   }
@@ -199,7 +193,7 @@ public void testReduceDims_Operand_Operand() {
       AtomicInteger index = new AtomicInteger();
       int[] expected = {8};
       reducedShape
-          .data()
+          .asTensor()
           .scalars()
           .forEach(
               s -> {
@@ -224,7 +218,7 @@ public void testReduceDims_Shape_Operand() {
       AtomicInteger index = new AtomicInteger();
       int[] expected1 = {8};
       reducedShape
-          .data()
+          .asTensor()
           .scalars()
           .forEach(
               s -> {
@@ -237,7 +231,7 @@ public void testReduceDims_Shape_Operand() {
       index.set(0);
       int[] expected2 = {2, 4};
       reducedShape
-          .data()
+          .asTensor()
           .scalars()
           .forEach(
               s -> {
@@ -250,7 +244,7 @@ public void testReduceDims_Shape_Operand() {
       index.set(0);
       int[] expected3 = {2, 2, 2};
       reducedShape
-          .data()
+          .asTensor()
           .scalars()
           .forEach(
               s -> {
@@ -274,10 +268,10 @@ public void testSqueeze() {
       Operand<TInt32> squeezed = Shapes.squeeze(scope, tfshape);
       AtomicInteger index = new AtomicInteger();
       int[] expected = {4, 2};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(squeezed.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(squeezed.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -301,10 +295,10 @@ public void testHead() {
       Operand<TInt32> head = Shapes.head(scope, tfshape);
       AtomicInteger index = new AtomicInteger();
       int[] expected = {4};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(head.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(head.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -328,10 +322,10 @@ public void testTake() {
       Operand<TInt32> take = Shapes.take(scope, tfshape, Constant.scalarOf(scope, 2));
       AtomicInteger index = new AtomicInteger();
       int[] expected = {4, 1};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(take.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(take.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -355,10 +349,10 @@ public void testTail() {
       Operand<TInt32> tail = Shapes.tail(scope, tfshape);
       AtomicInteger index = new AtomicInteger();
       int[] expected = {1};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(tail.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(tail.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -382,10 +376,10 @@ public void testTakeLast() {
       Operand<TInt32> takeLast = Shapes.takeLast(scope, tfshape, Constant.scalarOf(scope, 3));
       AtomicInteger index = new AtomicInteger();
       int[] expected = {1, 2, 1};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(takeLast.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(takeLast.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -408,10 +402,10 @@ public void testPrependInt() {
       Operand<TInt32> prepend = Shapes.prepend(scope, tfshape, 3);
       AtomicInteger index = new AtomicInteger();
       int[] expected = {3, 4, 2};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(prepend.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(prepend.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -429,15 +423,15 @@ public void testPrependLong() {
       Scope scope = new Scope(g);
       Operand operand = Constant.arrayOf(scope, new float[] {1, 2, 3, 4, 5, 6, 7, 8});
       Operand actual = Reshape.create(scope, operand, Constant.vectorOf(scope, new long[] {4, 2}));
-      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.DTYPE);
+      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.class);
 
       Operand<TInt64> prepend = Shapes.prepend(scope, tfshape, 1L);
       AtomicInteger index = new AtomicInteger();
       long[] expected = {1, 4, 2};
-      try (Tensor<TInt64> result =
-          session.runner().fetch(prepend.asOutput()).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          session.runner().fetch(prepend.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -465,10 +459,10 @@ public void testPrependShapeTInt32() {
       Operand<TInt32> prepend = Shapes.prepend(scope, tfshape1, tfshape2);
       AtomicInteger index = new AtomicInteger();
       int[] expected = {2, 4, 4, 2};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(prepend.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(prepend.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -490,16 +484,16 @@ public void testPrependShapeTInt64() {
       Operand operand2 = Constant.arrayOf(scope, new float[] {1, 2, 3, 4, 5, 6, 7, 8});
       Operand actual2 =
           Reshape.create(scope, operand2, Constant.vectorOf(scope, new long[] {2, 4}));
-      Shape<TInt64> tfshape1 = Shape.create(scope, actual1, TInt64.DTYPE);
-      Shape<TInt64> tfshape2 = Shape.create(scope, actual2, TInt64.DTYPE);
+      Shape<TInt64> tfshape1 = Shape.create(scope, actual1, TInt64.class);
+      Shape<TInt64> tfshape2 = Shape.create(scope, actual2, TInt64.class);
 
       Operand<TInt64> prepend = Shapes.prepend(scope, tfshape1, tfshape2);
       AtomicInteger index = new AtomicInteger();
       long[] expected = {2, 4, 4, 2};
-      try (Tensor<TInt64> result =
-          session.runner().fetch(prepend.asOutput()).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          session.runner().fetch(prepend.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -517,15 +511,15 @@ public void testAppendLong() {
       Scope scope = new Scope(g);
       Operand operand = Constant.arrayOf(scope, new float[] {1, 2, 3, 4, 5, 6, 7, 8});
       Operand actual = Reshape.create(scope, operand, Constant.vectorOf(scope, new long[] {4, 2}));
-      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.DTYPE);
+      Shape<TInt64> tfshape = Shape.create(scope, actual, TInt64.class);
 
       Operand<TInt64> append = Shapes.append(scope, tfshape, 2L);
       AtomicInteger index = new AtomicInteger();
       long[] expected = {4L, 2L, 2L};
-      try (Tensor<TInt64> result =
-          session.runner().fetch(append.asOutput()).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          session.runner().fetch(append.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -548,10 +542,10 @@ public void testAppendInt() {
       Operand<TInt32> append = Shapes.append(scope, tfshape, 2);
       AtomicInteger index = new AtomicInteger();
       int[] expected = {4, 2, 2};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(append.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(append.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -579,10 +573,10 @@ public void testAppendShapeTInt32() {
       Operand<TInt32> append = Shapes.append(scope, tfshape1, tfshape2);
       AtomicInteger index = new AtomicInteger();
       int[] expected = {4, 2, 2, 4};
-      try (Tensor<TInt32> result =
-          session.runner().fetch(append.asOutput()).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          session.runner().fetch(append.asOutput()).run().single()) {
         result
-            .data()
+
             .scalars()
             .forEach(
                 s -> {
@@ -604,16 +598,15 @@ public void testAppendShapeTInt64() {
       Operand operand2 = Constant.arrayOf(scope, new float[] {1, 2, 3, 4, 5, 6, 7, 8});
       Operand actual2 =
           Reshape.create(scope, operand2, Constant.vectorOf(scope, new long[] {2, 4}));
-      Shape<TInt64> tfshape1 = Shape.create(scope, actual1, TInt64.DTYPE);
-      Shape<TInt64> tfshape2 = Shape.create(scope, actual2, TInt64.DTYPE);
+      Shape<TInt64> tfshape1 = Shape.create(scope, actual1, TInt64.class);
+      Shape<TInt64> tfshape2 = Shape.create(scope, actual2, TInt64.class);
 
       Operand<TInt64> append = Shapes.append(scope, tfshape1, tfshape2);
       AtomicInteger index = new AtomicInteger();
       long[] expected = {4, 2, 2, 4};
-      try (Tensor<TInt64> result =
-          session.runner().fetch(append.asOutput()).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          session.runner().fetch(append.asOutput()).run().single()) {
         result
-            .data()
             .scalars()
             .forEach(
                 s -> {
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ZerosTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ZerosTest.java
index 9600f8b38fc..f84c7ce5530 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ZerosTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/op/core/ZerosTest.java
@@ -19,12 +19,11 @@
 import static org.junit.jupiter.api.Assertions.assertFalse;
 import static org.junit.jupiter.api.Assertions.assertTrue;
 
-import java.util.List;
 import org.junit.jupiter.api.Test;
 import org.tensorflow.Graph;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.Scope;
+import org.tensorflow.util.TensorList;
 import org.tensorflow.types.TBool;
 import org.tensorflow.types.TFloat32;
 import org.tensorflow.types.TFloat64;
@@ -41,9 +40,9 @@ public void createIntZeros() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TInt32> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TInt32.DTYPE);
-      try (Tensor<TInt32> result = sess.runner().fetch(op).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(0, s.getInt()));
+      Zeros<TInt32> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TInt32.class);
+      try (TInt32 result = sess.runner().fetch(op).run().single()) {
+        result.scalars().forEach(s -> assertEquals(0, s.getInt()));
       }
     }
   }
@@ -54,9 +53,9 @@ public void createFloatZeros() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TFloat32> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TFloat32.DTYPE);
-      try (Tensor<TFloat32> result = sess.runner().fetch(op.asOutput()).run().get(0).expect(TFloat32.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(0.0f, s.getFloat(), 0));
+      Zeros<TFloat32> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TFloat32.class);
+      try (TFloat32 result = sess.runner().fetch(op.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(0.0f, s.getFloat(), 0));
       }
     }
   }
@@ -67,9 +66,9 @@ public void createDoubleZeros() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TFloat64> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TFloat64.DTYPE);
-      try (Tensor<TFloat64> result = sess.runner().fetch(op.asOutput()).run().get(0).expect(TFloat64.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(0.0f, s.getDouble(), 0));
+      Zeros<TFloat64> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TFloat64.class);
+      try (TFloat64 result = sess.runner().fetch(op.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(0.0f, s.getDouble(), 0));
       }
     }
   }
@@ -80,9 +79,9 @@ public void createLongZeros() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TInt64> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TInt64.DTYPE);
-      try (Tensor<TInt64> result = sess.runner().fetch(op.asOutput()).run().get(0).expect(TInt64.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(0L, s.getLong()));
+      Zeros<TInt64> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TInt64.class);
+      try (TInt64 result = sess.runner().fetch(op.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(0L, s.getLong()));
       }
     }
   }
@@ -93,9 +92,9 @@ public void createBooleanZeros() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TBool> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TBool.DTYPE);
-      try (Tensor<TBool> result = sess.runner().fetch(op.asOutput()).run().get(0).expect(TBool.DTYPE)) {
-        result.data().scalars().forEach(s -> assertFalse(s.getBoolean()));
+      Zeros<TBool> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TBool.class);
+      try (TBool result = sess.runner().fetch(op.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertFalse(s.getBoolean()));
       }
    }
   }
@@ -106,9 +105,9 @@ public void createUint8Zeros() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TUint8> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TUint8.DTYPE);
-      try (Tensor<TUint8> result = sess.runner().fetch(op.asOutput()).run().get(0).expect(TUint8.DTYPE)) {
-        result.data().scalars().forEach(s -> assertEquals(0, s.getByte()));
+      Zeros<TUint8> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TUint8.class);
+      try (TUint8 result = sess.runner().fetch(op.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertEquals(0, s.getByte()));
       }
     }
   }
@@ -119,9 +118,9 @@ public void createStringZeros() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TString> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TString.DTYPE);
-      try (Tensor<TString> result = sess.runner().fetch(op.asOutput()).run().get(0).expect(TString.DTYPE)) {
-        result.data().scalars().forEach(s -> assertTrue(s.getObject().isEmpty()));
+      Zeros<TString> op = Zeros.create(scope, Constant.vectorOf(scope, shape), TString.class);
+      try (TString result = sess.runner().fetch(op.asOutput()).run().single()) {
+        result.scalars().forEach(s -> assertTrue(s.getObject().isEmpty()));
       }
     }
   }
@@ -132,8 +131,10 @@ public void operationsComposingZerosAreCorrectlyNamed() {
         Session sess = new Session(g)) {
       Scope scope = new Scope(g);
       long[] shape = {2, 2};
-      Zeros<TFloat32> zeros = Zeros.create(scope.withSubScope("test"), Constant.vectorOf(scope, shape), TFloat32.DTYPE);
-      List<Tensor<?>> results = sess.runner().addTarget("test/Zeros/Zero").addTarget("test/Zeros/Fill").run();
+      Zeros<TFloat32> zeros = Zeros.create(scope.withSubScope("test"), Constant.vectorOf(scope, shape), TFloat32.class);
+      try (TensorList results =
+          sess.runner().addTarget("test/Zeros/Zero").addTarget("test/Zeros/Fill").run()) {
+      }
     }
   }
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/NumericTypesTestBase.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/NumericTypesTestBase.java
index 87b24b0da2a..398b1693e14 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/NumericTypesTestBase.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/NumericTypesTestBase.java
@@ -21,90 +21,93 @@
 
 import org.junit.jupiter.api.Test;
 import org.tensorflow.EagerSession;
-import org.tensorflow.Tensor;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.index.Indices;
 import org.tensorflow.op.Ops;
 import org.tensorflow.op.core.Constant;
+import org.tensorflow.op.math.Add;
 import org.tensorflow.op.math.Sub;
-import org.tensorflow.ndarray.Shape;
-import org.tensorflow.ndarray.IntNdArray;
-import org.tensorflow.ndarray.NdArray;
-import org.tensorflow.ndarray.NdArrays;
-import org.tensorflow.ndarray.index.Indices;
 import org.tensorflow.types.family.TNumber;
 
-abstract class NumericTypesTestBase<T extends TNumber & NdArray<U>, U> {
+abstract class NumericTypesTestBase<T extends TNumber, U> {
 
   @Test
   public void initializeTensorsWithZeros() {
     // Allocate a tensor of 32-bits integer of the shape (2, 3, 2)
-    Tensor<T> tensor = allocateTensor(Shape.of(2, 3, 2));
-    NdArray<U> tensorData = tensor.data();
+    T tensor = allocateTensor(Shape.of(2, 3, 2));
 
-    assertEquals(3, tensorData.rank());
-    assertEquals(12, tensorData.size());
+    assertEquals(3, tensor.rank());
+    assertEquals(12, tensor.size());
+    NdArray<U> data = (NdArray<U>)tensor;
 
     try (EagerSession session = EagerSession.create()) {
       Ops tf = Ops.create(session);
 
       // Initialize tensor memory with zeros and take a snapshot
-      tensorData.scalars().forEach(scalar -> scalar.setObject(valueOf(0)));
+      data.scalars().forEach(scalar -> ((NdArray<U>)scalar).setObject(valueOf(0)));
       Constant<T> x = tf.constant(tensor);
 
       // Initialize the same tensor memory with ones and take a snapshot
-      tensorData.scalars().forEach(scalar -> scalar.setObject(valueOf(1)));
+      data.scalars().forEach(scalar -> ((NdArray<U>)scalar).setObject(valueOf(1)));
       Constant<T> y = tf.constant(tensor);
 
       // Subtract y from x and validate the result
       Sub<T> sub = tf.math.sub(x, y);
-      sub.data().scalars().forEach(scalar ->
+      ((NdArray<U>)sub.asTensor()).scalars().forEach(scalar ->
           assertEquals(valueOf(-1), scalar.getObject())
       );
     }
   }
 
   @Test
-  public void genericTest() {
-    IntNdArray heapData = NdArrays.vectorOf(0, 1, 2, 3);
+  public void setAndCompute() {
+    NdArray<U> heapData = allocateNdArray(Shape.of(4))
+        .setObject(valueOf(0), 0)
+        .setObject(valueOf(1), 1)
+        .setObject(valueOf(2), 2)
+        .setObject(valueOf(3), 3);
 
     // Creates a 2x2 matrix
-    try (Tensor<TInt32> tensor = TInt32.tensorOf(Shape.of(2, 2))) {
-      IntNdArray tensorData = tensor.data();
+    try (T tensor = allocateTensor(Shape.of(2, 2))) {
+      NdArray<U> data = (NdArray<U>)tensor;
 
       // Copy first 2 values of the vector to the first row of the matrix
-      tensorData.set(heapData.slice(Indices.range(0, 2)), 0);
+      data.set(heapData.slice(Indices.range(0, 2)), 0);
 
       // Copy values at an odd position in the vector as the second row of the matrix
-      tensorData.set(heapData.slice(Indices.odd()), 1);
+      data.set(heapData.slice(Indices.odd()), 1);
 
-      assertEquals(0, tensorData.getInt(0, 0));
-      assertEquals(1, tensorData.getInt(0, 1));
-      assertEquals(1, tensorData.getInt(1, 0));
-      assertEquals(3, tensorData.getInt(1, 1));
+      assertEquals(valueOf(0), data.getObject(0, 0));
+      assertEquals(valueOf(1), data.getObject(0, 1));
+      assertEquals(valueOf(1), data.getObject(1, 0));
+      assertEquals(valueOf(3), data.getObject(1, 1));
 
       // Read rows of the tensor in reverse order
-      IntNdArray reversedTensorData = tensorData.slice(Indices.all(), Indices.flip());
+      NdArray<U> flippedData = data.slice(Indices.flip(), Indices.flip());
 
-      assertEquals(1, reversedTensorData.getInt(0, 0));
-      assertEquals(0, reversedTensorData.getInt(0, 1));
-      assertEquals(3, reversedTensorData.getInt(1, 0));
-      assertEquals(1, reversedTensorData.getInt(1, 1));
+      assertEquals(valueOf(3), flippedData.getObject(0, 0));
+      assertEquals(valueOf(1), flippedData.getObject(0, 1));
+      assertEquals(valueOf(1), flippedData.getObject(1, 0));
+      assertEquals(valueOf(0), flippedData.getObject(1, 1));
 
       try (EagerSession session = EagerSession.create()) {
         Ops tf = Ops.create(session);
 
-        // Compute the power of the tensor by itself
-        Constant<TInt32> x = tf.constant(tensor);
-        IntNdArray result = tf.math.pow(x, x).data();
+        Add<T> add = tf.math.add(tf.constant(tensor), tf.constant(tensor));
+        NdArray<U> result = (NdArray<U>)add.asTensor();
 
-        // Validate result by computing the same operation in Java
-        tensorData.scalars().forEachIndexed((coords, s) ->
-          assertEquals(Math.pow(s.getInt(), s.getInt()), result.getInt(coords), 1e-7f)
-        );
+        assertEquals(valueOf(0), result.getObject(0, 0));
+        assertEquals(valueOf(2), result.getObject(0, 1));
+        assertEquals(valueOf(2), result.getObject(1, 0));
+        assertEquals(valueOf(6), result.getObject(1, 1));
       }
     }
   }
 
-  abstract Tensor<T> allocateTensor(Shape shape);
+  abstract T allocateTensor(Shape shape);
+
+  abstract NdArray<U> allocateNdArray(Shape shape);
 
   abstract U valueOf(Integer value);
 }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TBfloat16Test.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TBfloat16Test.java
index 8681e805e3d..17a6e0dd2b5 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TBfloat16Test.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TBfloat16Test.java
@@ -17,16 +17,22 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.Tensor;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 
 public class TBfloat16Test extends NumericTypesTestBase<TBfloat16, Float> {
 
   @Override
-  Tensor<TBfloat16> allocateTensor(Shape shape) {
+  TBfloat16 allocateTensor(Shape shape) {
     return TBfloat16.tensorOf(shape);
   }
 
+  @Override
+  NdArray<Float> allocateNdArray(Shape shape) {
+    return NdArrays.ofFloats(shape);
+  }
+
   @Override
   Float valueOf(Integer value) {
     return value.floatValue();
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat16Test.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat16Test.java
index b72fe6fc01c..c1ae8ad3b6d 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat16Test.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat16Test.java
@@ -17,16 +17,22 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.Tensor;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 
 public class TFloat16Test extends NumericTypesTestBase<TFloat16, Float> {
 
   @Override
-  Tensor<TFloat16> allocateTensor(Shape shape) {
+  TFloat16 allocateTensor(Shape shape) {
     return TFloat16.tensorOf(shape);
   }
 
+  @Override
+  NdArray<Float> allocateNdArray(Shape shape) {
+    return NdArrays.ofFloats(shape);
+  }
+
   @Override
   Float valueOf(Integer value) {
     return value.floatValue();
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat32Test.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat32Test.java
index c4b1f6023f3..8df96f2871a 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat32Test.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat32Test.java
@@ -17,16 +17,22 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.Tensor;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 
 public class TFloat32Test extends NumericTypesTestBase<TFloat32, Float> {
 
   @Override
-  Tensor<TFloat32> allocateTensor(Shape shape) {
+  TFloat32 allocateTensor(Shape shape) {
     return TFloat32.tensorOf(shape);
   }
 
+  @Override
+  NdArray<Float> allocateNdArray(Shape shape) {
+    return NdArrays.ofFloats(shape);
+  }
+
   @Override
   Float valueOf(Integer value) {
     return value.floatValue();
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat64Test.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat64Test.java
index 0e9c8947d0f..47b4b6d936a 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat64Test.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TFloat64Test.java
@@ -17,16 +17,22 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.Tensor;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 
 public class TFloat64Test extends NumericTypesTestBase<TFloat64, Double> {
 
   @Override
-  Tensor<TFloat64> allocateTensor(Shape shape) {
+  TFloat64 allocateTensor(Shape shape) {
     return TFloat64.tensorOf(shape);
   }
 
+  @Override
+  NdArray<Double> allocateNdArray(Shape shape) {
+    return NdArrays.ofDoubles(shape);
+  }
+
   @Override
   Double valueOf(Integer value) {
     return value.doubleValue();
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt32Test.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt32Test.java
index c52394bf210..9ea7f952f04 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt32Test.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt32Test.java
@@ -17,16 +17,24 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.Tensor;
+import static org.junit.jupiter.api.Assertions.assertEquals;
+
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 
 public class TInt32Test extends NumericTypesTestBase<TInt32, Integer> {
 
   @Override
-  Tensor<TInt32> allocateTensor(Shape shape) {
+  TInt32 allocateTensor(Shape shape) {
     return TInt32.tensorOf(shape);
   }
 
+  @Override
+  NdArray<Integer> allocateNdArray(Shape shape) {
+    return NdArrays.ofInts(shape);
+  }
+
   @Override
   Integer valueOf(Integer value) {
     return value;
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt64Test.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt64Test.java
index 261ac546fc5..a88f3fb4d6d 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt64Test.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TInt64Test.java
@@ -17,16 +17,22 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.Tensor;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 
 public class TInt64Test extends NumericTypesTestBase<TInt64, Long> {
 
   @Override
-  Tensor<TInt64> allocateTensor(Shape shape) {
+  TInt64 allocateTensor(Shape shape) {
     return TInt64.tensorOf(shape);
   }
 
+  @Override
+  NdArray<Long> allocateNdArray(Shape shape) {
+    return NdArrays.ofLongs(shape);
+  }
+
   @Override
   Long valueOf(Integer value) {
     return value.longValue();
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TStringTest.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TStringTest.java
index a4700aa652f..08048947648 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TStringTest.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TStringTest.java
@@ -23,7 +23,6 @@
 
 import java.nio.charset.StandardCharsets;
 import org.junit.jupiter.api.Test;
-import org.tensorflow.Tensor;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.ndarray.NdArray;
 import org.tensorflow.ndarray.NdArrays;
@@ -32,37 +31,28 @@ public class TStringTest {
 
   @Test
   public void createScalar() {
-    Tensor<TString> tensor = TString.scalarOf("Pretty vacant");
+    TString tensor = TString.scalarOf("Pretty vacant");
     assertNotNull(tensor);
-
-    TString data = tensor.data();
-    assertNotNull(data);
-    assertEquals(Shape.scalar(), data.shape());
-    assertEquals("Pretty vacant", data.getObject());
+    assertEquals(Shape.scalar(), tensor.shape());
+    assertEquals("Pretty vacant", tensor.getObject());
   }
 
     @Test
     public void createrScalarLongerThan127() {
-        Tensor<TString> tensor = TString.scalarOf("Long String 1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890 !");
+        TString tensor = TString.scalarOf("Long String 1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890 !");
         assertNotNull(tensor);
-
-        TString data = tensor.data();
-        assertNotNull(data);
-        assertEquals(Shape.scalar(), data.shape());
-        assertEquals("Long String 1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890 !", data.getObject());
+        assertEquals(Shape.scalar(), tensor.shape());
+        assertEquals("Long String 1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890 !", tensor.getObject());
     }
 
 
     @Test
   public void createVector() {
-    Tensor<TString> tensor = TString.vectorOf("Pretty", "vacant");
+    TString tensor = TString.vectorOf("Pretty", "vacant");
     assertNotNull(tensor);
-
-    TString data = tensor.data();
-    assertNotNull(data);
-    assertEquals(Shape.of(2), data.shape());
-    assertEquals("Pretty", data.getObject(0));
-    assertEquals("vacant", data.getObject(1));
+    assertEquals(Shape.of(2), tensor.shape());
+    assertEquals("Pretty", tensor.getObject(0));
+    assertEquals("vacant", tensor.getObject(1));
   }
 
   @Test
@@ -73,30 +63,27 @@ public void createCopy() {
         .setObject("New", 1, 0)
         .setObject("York", 1, 1);
 
-    Tensor<TString> tensor = TString.tensorOf(strings);
+    TString tensor = TString.tensorOf(strings);
     assertNotNull(tensor);
-
-    TString data = tensor.data();
-    assertNotNull(data);
     strings.scalars().forEachIndexed((idx, s) ->
-        assertEquals(s.getObject(), data.getObject(idx))
+        assertEquals(s.getObject(), tensor.getObject(idx))
     );
   }
 
   @Test
   public void defaultCharsetIsUtf8() {
-    Tensor<TString> tensor = TString.tensorOf(NdArrays.scalarOfObject(BABY_CHICK));
-    byte[] bytes = tensor.data().asBytes().getObject();
+    TString tensor = TString.tensorOf(NdArrays.scalarOfObject(BABY_CHICK));
+    byte[] bytes = tensor.asBytes().getObject();
     assertArrayEquals(new byte[] { (byte)0xF0, (byte)0x9F, (byte)0x90, (byte)0xA5 }, bytes);
-    assertEquals(BABY_CHICK, tensor.data().getObject());
+    assertEquals(BABY_CHICK, tensor.getObject());
   }
 
   @Test
   public void usingDifferentCharset() {
-    Tensor<TString> tensor = TString.tensorOf(StandardCharsets.UTF_16LE, NdArrays.scalarOfObject(BABY_CHICK));
-    byte[] bytes = tensor.data().asBytes().getObject();
+    TString tensor = TString.tensorOf(StandardCharsets.UTF_16LE, NdArrays.scalarOfObject(BABY_CHICK));
+    byte[] bytes = tensor.asBytes().getObject();
     assertArrayEquals(new byte[] { (byte)0x3D, (byte)0xD8, (byte)0x25, (byte)0xDC }, bytes);
-    assertEquals(BABY_CHICK, tensor.data().using(StandardCharsets.UTF_16LE).getObject());
+    assertEquals(BABY_CHICK, tensor.using(StandardCharsets.UTF_16LE).getObject());
   }
 
   @Test
@@ -106,11 +93,11 @@ public void initializingTensorWithRawBytes() {
     for (int i = 0; i < strings.length; ++i) {
       bytes.setObject(strings[i].getBytes(), i);
     }
-    Tensor<TString> tensor = TString.tensorOfBytes(bytes);
+    TString tensor = TString.tensorOfBytes(bytes);
     assertNotNull(tensor);
     assertEquals(bytes.shape(), tensor.shape());
 
-    NdArray<byte[]> tensorBytes = tensor.data().asBytes();
+    NdArray<byte[]> tensorBytes = tensor.asBytes();
     for (int i = 0; i < strings.length; ++i) {
       assertArrayEquals(bytes.getObject(i), tensorBytes.getObject(i));
     }
diff --git a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TUint8Test.java b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TUint8Test.java
index cc83087e018..ce7397d5878 100644
--- a/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TUint8Test.java
+++ b/tensorflow-core/tensorflow-core-api/src/test/java/org/tensorflow/types/TUint8Test.java
@@ -17,16 +17,22 @@
 
 package org.tensorflow.types;
 
-import org.tensorflow.Tensor;
+import org.tensorflow.ndarray.NdArray;
+import org.tensorflow.ndarray.NdArrays;
 import org.tensorflow.ndarray.Shape;
 
 public class TUint8Test extends NumericTypesTestBase<TUint8, Byte> {
 
   @Override
-  Tensor<TUint8> allocateTensor(Shape shape) {
+  TUint8 allocateTensor(Shape shape) {
     return TUint8.tensorOf(shape);
   }
 
+  @Override
+  NdArray<Byte> allocateNdArray(Shape shape) {
+    return NdArrays.ofBytes(shape);
+  }
+
   @Override
   Byte valueOf(Integer value) {
     return value.byteValue();
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ELU.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ELU.java
index ae3d7e8c896..d421bc1155d 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ELU.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ELU.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.activations;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TBool;
@@ -89,7 +88,7 @@ public Operand<T> call(Operand<T> input) {
     Operand<T> result = tf.nn.elu(input);
     if (alpha == 1.0) return result;
     else {
-      DataType<T> dataType = input.asOutput().dataType();
+      Class<T> dataType = input.asOutput().type();
       Operand<T> y = tf.math.mul(result, tf.dtypes.cast(tf.constant(alpha), dataType));
       Operand<TBool> cond = tf.math.greater(result, tf.dtypes.cast(tf.constant(0), dataType));
       return tf.select(cond, result, y);
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/HardSigmoid.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/HardSigmoid.java
index a486cbdc601..9e254a8fa34 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/HardSigmoid.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/HardSigmoid.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.activations;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.family.TFloating;
@@ -63,7 +62,7 @@ public HardSigmoid(Ops tf) {
    */
   @Override
   public Operand<T> call(Operand<T> input) {
-    DataType<T> dataType = input.asOutput().dataType();
+    Class<T> dataType = input.asOutput().type();
     Operand<T> point2 = tf.dtypes.cast(tf.constant(0.2), dataType);
     Operand<T> point5 = tf.dtypes.cast(tf.constant(0.5), dataType);
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ReLU.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ReLU.java
index c24cf71077d..d405aa1bc7d 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ReLU.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/activations/ReLU.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.activations;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.op.math.Greater;
@@ -99,7 +98,7 @@ public ReLU(Ops tf, float alpha, float maxValue, float threshold) {
   @Override
   public Operand<T> call(Operand<T> input) {
 
-    DataType<T> dataType = input.asOutput().dataType();
+    Class<T> dataType = input.asOutput().type();
 
     boolean clipMax = !Float.isNaN(maxValue);
     Operand<T> negativePart = null;
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/Dataset.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/Dataset.java
index 007bcb01a40..c1eca6aeb29 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/Dataset.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/Dataset.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.data;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.data.impl.BatchDataset;
 import org.tensorflow.framework.data.impl.MapDataset;
@@ -33,6 +32,7 @@
 import java.util.Iterator;
 import java.util.List;
 import java.util.function.Function;
+import org.tensorflow.types.family.TType;
 
 /**
  * Represents a potentially large list of independent elements (samples), and allows iteration and
@@ -41,11 +41,11 @@
 public abstract class Dataset implements Iterable<List<Operand<?>>> {
   protected Ops tf;
   private Operand<?> variant;
-  private List<DataType<?>> outputTypes;
+  private List<Class<? extends TType>> outputTypes;
   private List<Shape> outputShapes;
 
   public Dataset(
-      Ops tf, Operand<?> variant, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      Ops tf, Operand<?> variant, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     if (tf == null) {
       throw new IllegalArgumentException("Ops accessor cannot be null.");
     }
@@ -266,7 +266,7 @@ public DatasetIterator makeOneShotIterator() {
    * @return A new `Dataset`
    */
   public static Dataset fromTensorSlices(
-      Ops tf, List<Operand<?>> tensors, List<DataType<?>> outputTypes) {
+      Ops tf, List<Operand<?>> tensors, List<Class<? extends TType>> outputTypes) {
     return new TensorSliceDataset(tf, tensors, outputTypes);
   }
 
@@ -288,7 +288,7 @@ public Operand<?> getVariant() {
   }
 
   /** Get a list of output types for each component of this dataset. */
-  public List<DataType<?>> getOutputTypes() {
+  public List<Class<? extends TType>> getOutputTypes() {
     return this.outputTypes;
   }
 
@@ -305,7 +305,7 @@ public Ops getOpsInstance() {
   public String toString() {
     return "Dataset{"
         + "outputTypes="
-        + Arrays.toString(getOutputTypes().stream().map(DataType::name).toArray())
+        + Arrays.toString(getOutputTypes().stream().map(Class::getSimpleName).toArray())
         + ", outputShapes="
         + Arrays.toString(getOutputShapes().stream().map(Shape::toString).toArray())
         + "}";
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetIterator.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetIterator.java
index f4c4b681715..5dbb2bfacff 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetIterator.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetIterator.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.data;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Graph;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Op;
@@ -25,6 +24,7 @@
 import java.util.ArrayList;
 import java.util.Iterator;
 import java.util.List;
+import org.tensorflow.types.family.TType;
 
 /**
  * Represents the state of an iteration through a tf.data Datset. DatasetIterator is not a
@@ -106,7 +106,7 @@ public class DatasetIterator implements Iterable<List<Operand<?>>> {
   private Operand<?> iteratorResource;
   private Op initializer;
 
-  protected List<DataType<?>> outputTypes;
+  protected List<Class<? extends TType>> outputTypes;
   protected List<Shape> outputShapes;
 
   /**
@@ -124,7 +124,7 @@ public DatasetIterator(
       Ops tf,
       Operand<?> iteratorResource,
       Op initializer,
-      List<DataType<?>> outputTypes,
+      List<Class<? extends TType>> outputTypes,
       List<Shape> outputShapes) {
 
     this.tf = tf;
@@ -137,7 +137,7 @@ public DatasetIterator(
   public DatasetIterator(
       Ops tf,
       Operand<?> iteratorResource,
-      List<DataType<?>> outputTypes,
+      List<Class<? extends TType>> outputTypes,
       List<Shape> outputShapes) {
     this.tf = tf;
     this.iteratorResource = iteratorResource;
@@ -236,7 +236,7 @@ public Op makeInitializer(Dataset dataset) {
    * @return A new DatasetIterator
    */
   public static DatasetIterator fromStructure(
-      Ops tf, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      Ops tf, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     Operand<?> iteratorResource =
         tf.scope().env() instanceof Graph
             ? tf.data.iterator(EMPTY_SHARED_NAME, "", outputTypes, outputShapes)
@@ -272,7 +272,7 @@ public Iterator<List<Operand<?>>> iterator() {
 
       @Override
       public boolean hasNext() {
-        return nextOptional.hasValue().data().getBoolean();
+        return nextOptional.hasValue().asTensor().getBoolean();
       }
 
       @Override
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetOptional.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetOptional.java
index 925252c7298..6617c33eaf7 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetOptional.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/DatasetOptional.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.data;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.ndarray.Shape;
@@ -23,6 +22,7 @@
 
 import java.util.ArrayList;
 import java.util.List;
+import org.tensorflow.types.family.TType;
 
 /**
  * An optional represents the result of a dataset getNext operation that may fail, when the end of
@@ -36,11 +36,11 @@ public Operand<?> getOptionalVariant() {
   }
 
   private Operand<?> optionalVariant;
-  private List<DataType<?>> outputTypes;
+  private List<Class<? extends TType>> outputTypes;
   private List<Shape> outputShapes;
 
   public DatasetOptional(
-      Ops tf, Operand<?> optionalVariant, List<DataType<?>> outputTypes, List<Shape> outputShapes) {
+      Ops tf, Operand<?> optionalVariant, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes) {
     this.tf = tf;
     this.optionalVariant = optionalVariant;
     this.outputTypes = outputTypes;
@@ -75,7 +75,7 @@ public List<Operand<?>> getValue() {
   public static DatasetOptional fromComponents(
       Ops tf,
       List<Operand<?>> components,
-      List<DataType<?>> outputTypes,
+      List<Class<? extends TType>> outputTypes,
       List<Shape> outputShapes) {
     Operand<?> optionalVariant = tf.data.optionalFromValue(components);
     return new DatasetOptional(tf, optionalVariant, outputTypes, outputShapes);
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/BatchDataset.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/BatchDataset.java
index 277b049cf6f..f0561b2e61e 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/BatchDataset.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/BatchDataset.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.data.impl;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.data.Dataset;
 import org.tensorflow.op.Ops;
@@ -25,6 +24,7 @@
 import org.tensorflow.types.TInt64;
 
 import java.util.List;
+import org.tensorflow.types.family.TType;
 
 public class BatchDataset extends Dataset {
   public BatchDataset(
@@ -32,7 +32,7 @@ public BatchDataset(
       Operand<?> variant,
       Constant<TInt64> batchSize,
       Constant<TBool> dropRemainder,
-      List<DataType<?>> outputTypes,
+      List<Class<? extends TType>> outputTypes,
       List<Shape> outputShapes) {
     super(
         tf,
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/SkipDataset.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/SkipDataset.java
index 6731bac60b3..63b4208480b 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/SkipDataset.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/SkipDataset.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.data.impl;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.data.Dataset;
 import org.tensorflow.op.Ops;
@@ -24,6 +23,7 @@
 import org.tensorflow.types.TInt64;
 
 import java.util.List;
+import org.tensorflow.types.family.TType;
 
 public class SkipDataset extends Dataset {
 
@@ -31,7 +31,7 @@ public SkipDataset(
       Ops tf,
       Operand<?> variant,
       Constant<TInt64> count,
-      List<DataType<?>> outputTypes,
+      List<Class<? extends TType>> outputTypes,
       List<Shape> outputShapes) {
     super(
         tf,
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TFRecordDataset.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TFRecordDataset.java
index ed721b13ebf..00297152e90 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TFRecordDataset.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TFRecordDataset.java
@@ -34,7 +34,7 @@ public TFRecordDataset(
     super(
         tf,
         tf.data.tfRecordDataset(filenames, compressionType, bufferSize),
-        Collections.singletonList(TString.DTYPE),
+        Collections.singletonList(TString.class),
         Collections.singletonList(Shape.scalar()));
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TakeDataset.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TakeDataset.java
index 08c57d44a73..39ca9759e74 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TakeDataset.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TakeDataset.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.data.impl;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.data.Dataset;
 import org.tensorflow.op.Ops;
@@ -24,6 +23,7 @@
 import org.tensorflow.types.TInt64;
 
 import java.util.List;
+import org.tensorflow.types.family.TType;
 
 public class TakeDataset extends Dataset {
 
@@ -31,7 +31,7 @@ public TakeDataset(
       Ops tf,
       Operand<?> variant,
       Constant<TInt64> count,
-      List<DataType<?>> outputTypes,
+      List<Class<? extends TType>> outputTypes,
       List<Shape> outputShapes) {
     super(
         tf,
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TensorSliceDataset.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TensorSliceDataset.java
index 14405ebdaf5..eac0f1cc1ed 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TensorSliceDataset.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TensorSliceDataset.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.data.impl;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.data.Dataset;
 import org.tensorflow.op.Ops;
@@ -23,10 +22,11 @@
 
 import java.util.List;
 import java.util.stream.Collectors;
+import org.tensorflow.types.family.TType;
 
 public class TensorSliceDataset extends Dataset {
 
-  public TensorSliceDataset(Ops tf, List<Operand<?>> components, List<DataType<?>> outputTypes) {
+  public TensorSliceDataset(Ops tf, List<Operand<?>> components, List<Class<? extends TType>> outputTypes) {
     super(tf, makeVariant(tf, components, outputTypes), outputTypes, outputShapes(components));
   }
 
@@ -35,7 +35,7 @@ private static List<Shape> outputShapes(List<Operand<?>> components) {
   }
 
   private static Operand<?> makeVariant(
-      Ops tf, List<Operand<?>> components, List<DataType<?>> outputTypes) {
+      Ops tf, List<Operand<?>> components, List<Class<? extends TType>> outputTypes) {
     if (!(components.size() == outputTypes.size())) {
       throw new IllegalArgumentException(
           "Lists `tensors` and `dtypes` must have the same number of elements.");
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TextLineDataset.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TextLineDataset.java
index 4ef47825211..c9a26304778 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TextLineDataset.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/data/impl/TextLineDataset.java
@@ -34,7 +34,7 @@ public TextLineDataset(
     super(
         tf,
         tf.data.textLineDataset(filenames, compressionType, bufferSize),
-        Collections.singletonList(TString.DTYPE),
+        Collections.singletonList(TString.class),
         Collections.singletonList(Shape.scalar()));
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Constant.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Constant.java
index b4544de9bd0..16aed3db6ed 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Constant.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Constant.java
@@ -14,10 +14,11 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
+import org.tensorflow.types.TBool;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
 /**
@@ -85,17 +86,17 @@ public Constant(Ops tf, boolean value) {
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!(dtype.isNumeric() || dtype.isBoolean())) {
-      throw new IllegalArgumentException("DataType must be numeric or boolean: " + dtype.name());
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
+    if (!TNumber.class.isAssignableFrom(type) && type != TBool.class) {
+      throw new IllegalArgumentException("DataType must be numeric or boolean: " + type.getSimpleName());
     }
     switch (valueType) {
       case LONG:
-        return tf.fill(dims, tf.dtypes.cast(tf.constant(longValue), dtype));
+        return tf.fill(dims, tf.dtypes.cast(tf.constant(longValue), type));
       case DOUBLE:
-        return tf.fill(dims, tf.dtypes.cast(tf.constant(doubleValue), dtype));
+        return tf.fill(dims, tf.dtypes.cast(tf.constant(doubleValue), type));
       default:
-        return tf.fill(dims, tf.dtypes.cast(tf.constant(booleanValue), dtype));
+        return tf.fill(dims, tf.dtypes.cast(tf.constant(booleanValue), type));
     }
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Glorot.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Glorot.java
index 3d5d37b91d3..fd6b8542003 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Glorot.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Glorot.java
@@ -16,6 +16,7 @@
 package org.tensorflow.framework.initializers;
 
 import org.tensorflow.op.Ops;
+import org.tensorflow.types.family.TFloating;
 import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
@@ -70,7 +71,7 @@
  * @see VarianceScaling.Distribution
  * @see <a href="http://proceedings.mlr.press/v9/glorot10a.html">Glorot et al., 2010</a>
  */
-public class Glorot<T extends TType, U extends TNumber> extends VarianceScaling<T, U> {
+public class Glorot<T extends TFloating> extends VarianceScaling<T> {
 
   public static final double SCALE = 1.0;
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/He.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/He.java
index ce99da80bf7..5c0f3bbc8cb 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/He.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/He.java
@@ -15,6 +15,7 @@
 package org.tensorflow.framework.initializers;
 
 import org.tensorflow.op.Ops;
+import org.tensorflow.types.family.TFloating;
 import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
@@ -66,7 +67,7 @@
  *     href="https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html">He
  *     et al., 2015</a>
  */
-public class He<T extends TType, U extends TNumber> extends VarianceScaling<T, U> {
+public class He<T extends TFloating> extends VarianceScaling<T> {
 
   public static final double SCALE = 2.0;
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Identity.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Identity.java
index 34e6cd790f4..9f462dcfc5c 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Identity.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Identity.java
@@ -14,12 +14,12 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.utils.ShapeUtils;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TFloating;
 import org.tensorflow.types.family.TType;
 
 /**
@@ -38,7 +38,7 @@
  *
  * @param <T> The TType for the call operation
  */
-public class Identity<T extends TType> extends BaseInitializer<T> {
+public class Identity<T extends TFloating> extends BaseInitializer<T> {
   public static final double GAIN_DEFAULT = 1.0;
 
   private final double gain;
@@ -66,10 +66,7 @@ public Identity(Ops tf, double gain) {
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!dtype.isFloating()) {
-      throw new IllegalArgumentException("DataType must be a float type: " + dtype.name());
-    }
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
     Shape shape = ShapeUtils.toShape(tf.scope(), dims);
     if (shape.numDimensions() != 2) {
       throw new IllegalArgumentException("2D matrix required, got " + shape.numDimensions());
@@ -79,9 +76,9 @@ public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
     Shape diagShape = Shape.of(diagSize);
 
     Operand<T> op;
-    Operand<T> zero = tf.dtypes.cast(tf.constant(0), dtype);
+    Operand<T> zero = tf.dtypes.cast(tf.constant(0), type);
     Operand<T> diagOnes =
-        tf.fill(tf.constant(diagShape.asArray()), tf.dtypes.cast(tf.constant(1.0), dtype));
+        tf.fill(tf.constant(diagShape.asArray()), tf.dtypes.cast(tf.constant(1.0), type));
     if (isSquare) {
       op =
           tf.linalg.matrixDiag(
@@ -91,10 +88,10 @@ public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
               tf.constant((int) shape.size(1)),
               zero);
     } else {
-      Operand<T> zeroMatrix = tf.zeros(dims, dtype);
+      Operand<T> zeroMatrix = tf.zeros(dims, type);
       op = tf.linalg.matrixSetDiag(zeroMatrix, diagOnes, tf.constant(0));
     }
 
-    return tf.math.mul(op, tf.dtypes.cast(tf.constant(gain), dtype));
+    return tf.math.mul(op, tf.dtypes.cast(tf.constant(gain), type));
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Initializer.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Initializer.java
index 59dce1fc02e..4beb218783b 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Initializer.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Initializer.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.family.TType;
@@ -30,8 +29,8 @@ public interface Initializer<T extends TType> {
    * Generates the operation used to perform the initialization.
    *
    * @param dims the shape dimensions
-   * @param dtype the data type
+   * @param type the type of tensor
    * @return An operand for the initialization.
    */
-  Operand<T> call(Operand<TInt64> dims, DataType<T> dtype);
+  Operand<T> call(Operand<TInt64> dims, Class<T> type);
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/LeCun.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/LeCun.java
index e2268412fc3..0f29f21284a 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/LeCun.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/LeCun.java
@@ -15,6 +15,7 @@
 package org.tensorflow.framework.initializers;
 
 import org.tensorflow.op.Ops;
+import org.tensorflow.types.family.TFloating;
 import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
@@ -76,7 +77,7 @@
  *     al., 1998</a>
  * @see VarianceScaling.Distribution
  */
-public class LeCun<T extends TType, U extends TNumber> extends VarianceScaling<T, U> {
+public class LeCun<T extends TFloating> extends VarianceScaling<T> {
 
   /**
    * Creates a LeCunNormal Initializer
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Ones.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Ones.java
index b78f34e3d35..ef33d37bcb0 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Ones.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Ones.java
@@ -14,10 +14,11 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
+import org.tensorflow.types.TBool;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
 /**
@@ -56,10 +57,10 @@ public Ones(Ops tf) {
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!(dtype.isNumeric() || dtype.isBoolean())) {
-      throw new IllegalArgumentException("DataType must be numeric or boolean: " + dtype.name());
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
+    if (!TNumber.class.isAssignableFrom(type) && type != TBool.class) {
+      throw new IllegalArgumentException("DataType must be numeric or boolean: " + type.getSimpleName());
     }
-    return tf.fill(dims, tf.dtypes.cast(tf.constant(1.0), dtype));
+    return tf.fill(dims, tf.dtypes.cast(tf.constant(1.0), type));
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Orthogonal.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Orthogonal.java
index 48e2c56d5be..1aada14d3c4 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Orthogonal.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Orthogonal.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.Output;
 import org.tensorflow.framework.utils.ShapeUtils;
@@ -22,6 +21,7 @@
 import org.tensorflow.op.Ops;
 import org.tensorflow.op.linalg.Qr;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TFloating;
 import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
@@ -50,7 +50,7 @@
  * @param <T> The TType for the call operation
  * @param <U> The TNumber for the call operation
  */
-public class Orthogonal<T extends TType, U extends TNumber> extends BaseInitializer<T> {
+public class Orthogonal<T extends TFloating> extends BaseInitializer<T> {
 
   public static final double GAIN_DEFAULT = 1.0;
 
@@ -84,10 +84,7 @@ public Orthogonal(Ops tf, double gain, long seed) {
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!dtype.isFloating()) {
-      throw new IllegalArgumentException("Expected floating point type, got " + dtype.name());
-    }
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
     Shape dimsShape = ShapeUtils.toShape(tf.scope(), dims);
     if (dimsShape.numDimensions() < 2) {
       throw new IllegalArgumentException(
@@ -100,22 +97,17 @@ public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
     long numCols = dimsShape.size(i);
     Shape flatShape = Shape.of(Math.max(numRows, numCols), Math.min(numRows, numCols));
     long[] seeds = {seed, 0};
-    @SuppressWarnings("unchecked")
-    DataType<U> numdType = (DataType<U>) dtype;
-    @SuppressWarnings("unchecked")
     Operand<T> op =
-        (Operand<T>)
-            tf.random.statelessRandomNormal(tf.constant(flatShape), tf.constant(seeds), numdType);
-
+        tf.random.statelessRandomNormal(tf.constant(flatShape), tf.constant(seeds), type);
     Qr.Options qrOptions = Qr.fullMatrices(false);
     Qr<T> qrOp = tf.linalg.qr(op, qrOptions);
     Output<T> qo = qrOp.q();
     Output<T> ro = qrOp.r();
     Operand<T> diagOp =
-        tf.linalg.matrixDiagPart(ro, tf.constant(0), tf.dtypes.cast(tf.constant(0), dtype));
+        tf.linalg.matrixDiagPart(ro, tf.constant(0), tf.dtypes.cast(tf.constant(0), type));
     Operand<T> qop = tf.math.mul(qo, tf.math.sign(diagOp));
     if (numRows < numCols) qop = tf.linalg.transpose(qop, null);
 
-    return tf.math.mul(qop, tf.dtypes.cast(tf.constant(this.gain), dtype));
+    return tf.math.mul(qop, tf.dtypes.cast(tf.constant(this.gain), type));
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomNormal.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomNormal.java
index f2d8a0d8e6e..5c2311d7439 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomNormal.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomNormal.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt64;
@@ -37,7 +36,7 @@
  * @param <T> The TType for the call operation
  * @param <U> The TNumber for the call operation
  */
-public class RandomNormal<T extends TType, U extends TNumber> extends BaseInitializer<T> {
+public class RandomNormal<T extends TNumber> extends BaseInitializer<T> {
   public static final double MEAN_DEFAULT = 0.0;
   public static final double STDDEV_DEFAULT = 1.0;
 
@@ -87,16 +86,10 @@ public RandomNormal(Ops tf, double mean, double stddev, long seed) {
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!dtype.isNumeric())
-      throw new IllegalArgumentException("The data type must be numeric. Found : " + dtype.name());
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
     long[] seeds = {seed, 0};
-    @SuppressWarnings("unchecked")
-    DataType<U> numdType = (DataType<U>) dtype;
-    @SuppressWarnings("unchecked")
-    Operand<T> distOp =
-        (Operand<T>) tf.random.statelessRandomNormal(dims, tf.constant(seeds), numdType);
-    Operand<T> op = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(this.stddev), dtype));
-    return tf.math.add(op, tf.dtypes.cast(tf.constant(mean), dtype));
+    Operand<T> distOp = tf.random.statelessRandomNormal(dims, tf.constant(seeds), type);
+    Operand<T> op = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(this.stddev), type));
+    return tf.math.add(op, tf.dtypes.cast(tf.constant(mean), type));
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomUniform.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomUniform.java
index b665729675d..77c1927bdfa 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomUniform.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/RandomUniform.java
@@ -14,11 +14,11 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.op.random.RandomUniformInt;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TInteger;
 import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
@@ -38,7 +38,7 @@
  * @param <T> The TType for the call operation
  * @param <U> The TNumber for the call operation
  */
-public class RandomUniform<T extends TType, U extends TNumber> extends BaseInitializer<T> {
+public class RandomUniform<T extends TNumber> extends BaseInitializer<T> {
 
   public static final double MINVAL_DEFAULT = -0.05;
   public static final double MAXVAL_DEFAULT = 0.05;
@@ -77,39 +77,28 @@ public RandomUniform(Ops tf, double minval, double maxval, long seed) {
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!dtype.isNumeric())
-      throw new IllegalArgumentException("The data type must be numeric. Found : " + dtype.name());
-    @SuppressWarnings("unchecked")
-    DataType<U> numdType = (DataType<U>) dtype;
-    Operand<U> distOp;
-
-    if (dtype.isInteger()) {
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
+    Operand<T> distOp;
+    if (TInteger.class.isAssignableFrom(type)) {
       RandomUniformInt.Options options = RandomUniformInt.seed(this.seed);
       distOp =
           tf.random.randomUniformInt(
               dims,
-              tf.dtypes.cast(tf.constant(this.minval), numdType),
-              tf.dtypes.cast(tf.constant(this.maxval), numdType),
+              tf.dtypes.cast(tf.constant(this.minval), type),
+              tf.dtypes.cast(tf.constant(this.maxval), type),
               options);
-      @SuppressWarnings("unchecked")
-      Operand<T> distOpT = (Operand<T>) distOp;
-      return distOpT;
     } else {
       long[] seeds = {seed, 0};
-      distOp = tf.random.statelessRandomUniform(dims, tf.constant(seeds), numdType);
-      @SuppressWarnings("unchecked")
-      Operand<T> distOpT = (Operand<T>) distOp;
+      distOp = tf.random.statelessRandomUniform(dims, tf.constant(seeds), type);
       if (this.minval == 0) {
         if (this.maxval != 1.0) {
-          distOpT = tf.math.mul(distOpT, tf.dtypes.cast(tf.constant(this.maxval), dtype));
+          distOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(this.maxval), type));
         }
       } else {
-        distOpT =
-            tf.math.mul(distOpT, tf.dtypes.cast(tf.constant(this.maxval - this.minval), dtype));
-        distOpT = tf.math.add(distOpT, tf.dtypes.cast(tf.constant(this.minval), dtype));
+        distOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(this.maxval - this.minval), type));
+        distOp = tf.math.add(distOp, tf.dtypes.cast(tf.constant(this.minval), type));
       }
-      return distOpT;
     }
+    return distOp;
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/TruncatedNormal.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/TruncatedNormal.java
index c71cf9a630e..c38c073f282 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/TruncatedNormal.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/TruncatedNormal.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt64;
@@ -37,7 +36,7 @@
  * @param <T> The TType for the call operation
  * @param <U> The TNumber for the call operation
  */
-public class TruncatedNormal<T extends TType, U extends TNumber> extends BaseInitializer<T> {
+public class TruncatedNormal<T extends TNumber> extends BaseInitializer<T> {
 
   public static final double MEAN_DEFAULT = 0.0;
   public static final double STDDEV_DEFAULT = 0.05;
@@ -76,17 +75,11 @@ public TruncatedNormal(Ops tf, double mean, double stddev, long seed) {
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!dtype.isNumeric())
-      throw new IllegalArgumentException("The data type must be numeric. Found : " + dtype.name());
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
     long[] seeds = {seed,0};
-    @SuppressWarnings("unchecked")
-    DataType<U> numdType = (DataType<U>) dtype;
-    Operand<U> distOp = tf.random.statelessTruncatedNormal(dims, tf.constant(seeds), numdType);
-    @SuppressWarnings("unchecked")
-    Operand<T> distOpT = (Operand<T>) distOp;
+    Operand<T> distOp = tf.random.statelessTruncatedNormal(dims, tf.constant(seeds), type);
     return tf.math.add(
-        tf.math.mul(distOpT, tf.dtypes.cast(tf.constant(stddev), dtype)),
-        tf.dtypes.cast(tf.constant(mean), dtype));
+        tf.math.mul(distOp, tf.dtypes.cast(tf.constant(stddev), type)),
+        tf.dtypes.cast(tf.constant(mean), type));
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/VarianceScaling.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/VarianceScaling.java
index fd33adadd5c..fdf8920ce1b 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/VarianceScaling.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/VarianceScaling.java
@@ -14,12 +14,12 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.utils.ShapeUtils;
 import org.tensorflow.ndarray.Shape;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt64;
+import org.tensorflow.types.family.TFloating;
 import org.tensorflow.types.family.TNumber;
 import org.tensorflow.types.family.TType;
 
@@ -58,7 +58,7 @@
  * @see VarianceScaling.Mode
  * @see VarianceScaling.Distribution
  */
-public class VarianceScaling<T extends TType, U extends TNumber> extends BaseInitializer<T> {
+public class VarianceScaling<T extends TFloating> extends BaseInitializer<T> {
 
   public static final double SCALE_DEFAULT = 1.0;
   public static final Mode MODE_DEFAULT = Mode.FAN_IN;
@@ -102,10 +102,7 @@ public VarianceScaling(Ops tf, double scale, Mode mode, Distribution distributio
 
   /** {@inheritDoc} */
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
-    if (!dtype.isFloating()) {
-      throw new IllegalArgumentException("Expected floating point type, got " + dtype.name());
-    }
+  public Operand<T> call(Operand<TInt64> dims, Class<T> type) {
     Shape shape = ShapeUtils.toShape(this.tf.scope(), dims);
     double lscale = this.scale;
     double[] fans /* fanIn, fanOut */ = computeFans(shape);
@@ -120,32 +117,28 @@ public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
         lscale /= Math.max(1., (fans[0] + fans[1]) / 2.);
         break;
     }
-    Operand<U> distOp;
-    Operand<U> mulOp = null;
-    @SuppressWarnings("unchecked")
-    DataType<U> numdType = (DataType<U>) dtype;
+    Operand<T> distOp;
+    Operand<T> mulOp = null;
     double stddev;
     long[] seeds = {seed, 0};
     switch (distribution) {
       case TRUNCATED_NORMAL:
-        distOp = tf.random.statelessTruncatedNormal(dims, tf.constant(seeds), numdType);
+        distOp = tf.random.statelessTruncatedNormal(dims, tf.constant(seeds), type);
         stddev = Math.sqrt(lscale) / .87962566103423978;
-        mulOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(stddev), numdType));
+        mulOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(stddev), type));
         break;
       case NORMAL:
-        distOp = tf.random.statelessRandomNormal(dims, tf.constant(seeds), numdType);
+        distOp = tf.random.statelessRandomNormal(dims, tf.constant(seeds), type);
         stddev = Math.sqrt(lscale);
-        mulOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(stddev), numdType));
+        mulOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(stddev), type));
         break;
       case UNIFORM:
-        distOp = tf.random.statelessRandomUniform(dims, tf.constant(seeds), numdType);
+        distOp = tf.random.statelessRandomUniform(dims, tf.constant(seeds), type);
         stddev = Math.sqrt(3.0 * lscale);
-        mulOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(stddev), numdType));
+        mulOp = tf.math.mul(distOp, tf.dtypes.cast(tf.constant(stddev), type));
         break;
     }
-
-    // Need to cast TNumber <U> to TType <T>
-    return (Operand<T>) mulOp;
+    return mulOp;
   }
 
   /**
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Zeros.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Zeros.java
index 09dd512ffaa..1582750b37b 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Zeros.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/initializers/Zeros.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.initializers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt64;
@@ -46,7 +45,7 @@ public Zeros(Ops tf) {
   }
 
   @Override
-  public Operand<T> call(Operand<TInt64> dims, DataType<T> dtype) {
+  public Operand<T> call(Operand<TInt64> dims, Class<T> dtype) {
     return tf.zeros(dims, dtype);
   }
 }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/BinaryCrossentropy.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/BinaryCrossentropy.java
index effdf990f71..88ab71eb4be 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/BinaryCrossentropy.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/BinaryCrossentropy.java
@@ -217,8 +217,8 @@ public <T extends TNumber, U extends TNumber> Operand<T> call(
               getTF(),
               "predictions range check [0-1]",
               predictions,
-              cast(getTF(), getTF().constant(0), predictions.asOutput().dataType()),
-              cast(getTF(), getTF().constant(1), predictions.asOutput().dataType()));
+              cast(getTF(), getTF().constant(0), predictions.asOutput().type()),
+              cast(getTF(), getTF().constant(1), predictions.asOutput().type()));
 
     } else {
       lPredictions = predictions;
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/CategoricalCrossentropy.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/CategoricalCrossentropy.java
index 7701ebfb806..d7bb4ab5128 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/CategoricalCrossentropy.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/CategoricalCrossentropy.java
@@ -256,8 +256,8 @@ public <T extends TNumber, U extends TNumber> Operand<T> call(
               getTF(),
               "predictions range check [0-1]",
               predictions,
-              cast(getTF(), getTF().constant(0), predictions.asOutput().dataType()),
-              cast(getTF(), getTF().constant(1), predictions.asOutput().dataType()));
+              cast(getTF(), getTF().constant(0), predictions.asOutput().type()),
+              cast(getTF(), getTF().constant(1), predictions.asOutput().type()));
 
     } else {
       lPredictions = predictions;
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Hinge.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Hinge.java
index 5fdfd4c9b96..b5bf2d32355 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Hinge.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Hinge.java
@@ -124,15 +124,15 @@ public Hinge(Ops tf, String name, Reduction reduction) {
   public <T extends TNumber, U extends TNumber> Operand<T> call(
       Operand<U> labels, Operand<T> predictions, Operand<T> sampleWeights) {
     @SuppressWarnings("unchecked")
-    Operand<T> tLabels = predictions.asOutput().dataType() == labels.asOutput().dataType() ?
+    Operand<T> tLabels = predictions.asOutput().type() == labels.asOutput().type() ?
             (Operand<T>)labels :
-            cast(tf,  labels, predictions.asOutput().dataType());
+            cast(tf,  labels, predictions.asOutput().type());
     tLabels = LossesHelper.valueCheck(
             getTF(),
             "labels value check [-1, 0, 1]",
             tLabels,
             cast(getTF(), getTF().constant(new int[] { -1, 0, 1}),
-                    predictions.asOutput().dataType()));
+                    predictions.asOutput().type()));
 
     Operand<T> losses = Losses.hinge(getTF(), tLabels, predictions);
     return LossesHelper.computeWeightedLoss(getTF(), losses, getReduction(), sampleWeights);
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Losses.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Losses.java
index 7a633ede2bf..1d5bbbf41e5 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Losses.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/Losses.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.losses;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.losses.impl.LossTuple;
 import org.tensorflow.framework.losses.impl.LossesHelper;
@@ -51,7 +50,7 @@ public class Losses {
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> meanAbsoluteError(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    Operand<T> tLabels = cast(tf, labels, predictions.asOutput().dataType());
+    Operand<T> tLabels = cast(tf, labels, predictions.asOutput().type());
     LossTuple<T> ops = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = ops.getTarget();
     tLabels = ops.getLabels();
@@ -73,7 +72,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> meanAbsoluteErro
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> meanSquaredError(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    Operand<T> tLabels = cast(tf, labels, predictions.asOutput().dataType());
+    Operand<T> tLabels = cast(tf, labels, predictions.asOutput().type());
     LossTuple<T> ops = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = ops.getTarget();
     tLabels = ops.getLabels();
@@ -94,7 +93,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> meanSquaredError
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> meanAbsolutePercentageError(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> ops = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = ops.getTarget();
@@ -121,7 +120,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> meanAbsolutePerc
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> meanSquaredLogarithmicError(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> ops = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = ops.getTarget();
@@ -152,7 +151,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> meanSquaredLogar
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> binaryCrossentropy(
       Ops tf, Operand<U> labels, Operand<T> predictions, boolean fromLogits, float labelSmoothing) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> ops = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = ops.getTarget();
@@ -193,7 +192,7 @@ private static <T extends TNumber> Operand<T> binaryCrossentropyHelper(
     }
     */
 
-    DataType<T> dataType = output.asOutput().dataType();
+    Class<T> dataType = output.asOutput().type();
     Operand<T> one = cast(tf, tf.constant(1), dataType);
     Operand<T> epsilonConst = cast(tf, tf.constant(EPSILON), dataType);
     Operand<T> oneMinusEpsilonConst = tf.math.sub(one, epsilonConst);
@@ -231,7 +230,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> categoricalCross
       boolean fromLogits,
       float labelSmoothing,
       int axis) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> ops = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = ops.getTarget();
@@ -284,7 +283,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> categoricalCross
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> categoricalHinge(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -330,7 +329,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> categoricalHinge
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> cosineSimilarity(
       Ops tf, Operand<U> labels, Operand<T> predictions, int axis) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -357,7 +356,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> cosineSimilarity
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> hinge(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -393,7 +392,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> hinge(
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> huber(
       Ops tf, Operand<U> labels, Operand<T> predictions, float delta) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -424,7 +423,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> huber(
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> kullbackLeiblerDivergence(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -454,7 +453,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> kullbackLeiblerD
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> logCosh(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -482,7 +481,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> logCosh(
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> poisson(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -509,7 +508,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> poisson(
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> sparseCategoricalCrossentropy(
       Ops tf, Operand<U> labels, Operand<T> predictions, boolean fromLogits, int axis) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> epsilonConst = cast(tf, tf.constant(EPSILON), dataType);
     Operand<T> one = cast(tf, tf.constant(1), dataType);
     Operand<T> oneMinusEpsilonConst = tf.math.sub(one, epsilonConst);
@@ -546,7 +545,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> sparseCategorica
       predictions = tf.linalg.transpose(predictions, tf.constant(axisNew));
     }
 
-    Operand<TInt64> iLabels = cast(tf, labels, TInt64.DTYPE);
+    Operand<TInt64> iLabels = cast(tf, labels, TInt64.class);
 
     // Try to adjust the shape so that rank of labels = rank of logits - 1.
     Shape labelsShape = labels.asOutput().shape();
@@ -586,7 +585,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> sparseCategorica
    */
   public static <T extends TNumber, U extends TNumber> Operand<T> squaredHinge(
       Ops tf, Operand<U> labels, Operand<T> predictions) {
-    DataType<T> dataType = predictions.asOutput().dataType();
+    Class<T> dataType = predictions.asOutput().type();
     Operand<T> tLabels = cast(tf, labels, dataType);
     LossTuple<T> lossTuple = LossesHelper.squeezeOrExpandDimensions(tf, tLabels, predictions, null);
     predictions = lossTuple.getTarget();
@@ -614,7 +613,7 @@ public static <T extends TNumber, U extends TNumber> Operand<T> squaredHinge(
    */
   private static <T extends TNumber> Operand<T> smoothBinaryLabels(
       Ops tf, Operand<T> labels, float labelSmoothing) {
-    DataType<T> dataType = labels.asOutput().dataType();
+    Class<T> dataType = labels.asOutput().type();
     Operand<T> oneMinusSmoothing = cast(tf, tf.constant(1.f - labelSmoothing), dataType);
     Operand<T> halfSmoothing = cast(tf, tf.constant(0.5F * labelSmoothing), dataType);
     return tf.math.add(tf.math.mul(labels, oneMinusSmoothing), halfSmoothing);
@@ -633,7 +632,7 @@ private static <T extends TNumber> Operand<T> smoothBinaryLabels(
    */
   private static <T extends TNumber> Operand<T> smoothCategoricalLabels(
       Ops tf, Operand<T> labels, float labelSmoothing) {
-    DataType<T> dataType = labels.asOutput().dataType();
+    Class<T> dataType = labels.asOutput().type();
     Operand<T> smoothing = cast(tf, tf.constant(labelSmoothing), dataType);
     Shape labelsShape = labels.asOutput().shape();
     int numDims = labelsShape.numDimensions();
@@ -656,7 +655,7 @@ public static <T extends TNumber> Operand<T> l2Normalize(Ops tf, Operand<T> x, i
         tf.reduceSum(tf.math.square(x), tf.constant(axis), ReduceSum.keepDims(Boolean.TRUE));
     Operand<T> invNorm =
         tf.math.rsqrt(
-            tf.math.maximum(squareSum, cast(tf, tf.constant(1e-12F), x.asOutput().dataType())));
+            tf.math.maximum(squareSum, cast(tf, tf.constant(1e-12F), x.asOutput().type())));
     return tf.math.mul(x, invNorm);
   }
 
@@ -669,7 +668,7 @@ public static <T extends TNumber> Operand<T> l2Normalize(Ops tf, Operand<T> x, i
    * @return the labels, possibly converted into -1/1.
    */
   private static <T extends TNumber> Operand<T> maybeConvertLabels(Ops tf, Operand<T> labels) {
-    DataType<T> dataType = labels.asOutput().dataType();
+    Class<T> dataType = labels.asOutput().type();
 
     Operand<T> one = cast(tf, tf.constant(1), dataType);
     Operand<T> zero = cast(tf, tf.constant(0), dataType);
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SparseCategoricalCrossentropy.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SparseCategoricalCrossentropy.java
index 5586a4da889..9372126d3ab 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SparseCategoricalCrossentropy.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SparseCategoricalCrossentropy.java
@@ -205,8 +205,8 @@ public <T extends TNumber, U extends TNumber> Operand<T> call(
                       getTF(),
                       "predictions range check [0-1]",
                       predictions,
-                      cast(getTF(), getTF().constant(0), predictions.asOutput().dataType()),
-                      cast(getTF(), getTF().constant(1), predictions.asOutput().dataType()));
+                      cast(getTF(), getTF().constant(0), predictions.asOutput().type()),
+                      cast(getTF(), getTF().constant(1), predictions.asOutput().type()));
 
     } else {
       lPredictions = predictions;
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SquaredHinge.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SquaredHinge.java
index 182ce592e55..9b712dcc712 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SquaredHinge.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/SquaredHinge.java
@@ -125,15 +125,15 @@ public SquaredHinge(Ops tf, String name, Reduction reduction) {
   public <T extends TNumber, U extends TNumber> Operand<T> call(
       Operand<U> labels, Operand<T> predictions, Operand<T> sampleWeights) {
     @SuppressWarnings("unchecked")
-    Operand<T> tLabels = predictions.asOutput().dataType() == labels.asOutput().dataType() ?
+    Operand<T> tLabels = predictions.asOutput().type() == labels.asOutput().type() ?
             (Operand<T>)labels :
-            cast(tf,  labels, predictions.asOutput().dataType());
+            cast(tf,  labels, predictions.asOutput().type());
     tLabels = LossesHelper.valueCheck(
             getTF(),
             "labels value check [-1, 0, 1]",
             tLabels,
             cast(getTF(), getTF().constant(new int[] { -1, 0, 1}),
-                    predictions.asOutput().dataType()));
+                    predictions.asOutput().type()));
     Operand<T> losses = Losses.squaredHinge(getTF(), tLabels, predictions);
     return LossesHelper.computeWeightedLoss(getTF(), losses, getReduction(), sampleWeights);
   }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/impl/LossesHelper.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/impl/LossesHelper.java
index 463296a1f50..1a9f1339e24 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/impl/LossesHelper.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/losses/impl/LossesHelper.java
@@ -14,7 +14,6 @@
 =======================================================================*/
 package org.tensorflow.framework.losses.impl;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.framework.losses.Reduction;
 import org.tensorflow.ndarray.Shape;
@@ -249,7 +248,7 @@ public static <T extends TNumber> LossTuple<T> removeSqueezableDimensions(
    */
   public static <T extends TNumber> Operand<T> computeWeightedLoss(
       Ops tf, Operand<T> loss, Reduction reduction, Operand<T> sampleWeight) {
-    DataType<T> dataType = loss.asOutput().dataType();
+    Class<T> dataType = loss.asOutput().type();
     if (sampleWeight == null) {
       sampleWeight = cast(tf, tf.constant(1), dataType);
     }
@@ -300,7 +299,7 @@ public static <T extends TNumber> Operand<T> safeMean(
       Ops tf, Operand<T> losses, long numElements) {
     Operand<T> totalLoss = tf.reduceSum(losses, allAxes(tf, losses));
     return tf.math.divNoNan(
-        totalLoss, cast(tf, tf.constant(numElements), losses.asOutput().dataType()));
+        totalLoss, cast(tf, tf.constant(numElements), losses.asOutput().type()));
   }
 
   /**
@@ -361,7 +360,7 @@ public static <T extends TNumber> Operand<T> rangeCheck(
           tf.withSubScope("rangeCheck")
               .withControlDependencies(Collections.singletonList(assertThat));
       return ltf.identity(values);
-    } else if (!cond.asOutput().data().getBoolean())
+    } else if (!cond.asOutput().asTensor().getBoolean())
       throw new IllegalArgumentException(String.format("%s : values out of range", prefix));
     else return values;
   }
@@ -386,7 +385,7 @@ public static <T extends TNumber> Operand<T> valueCheck(
       Ops tf, String prefix, Operand<T> values, Operand<T> allowedValues) {
     Operand<T> flatValues =
         tf.reshape(values, tf.constant(Shape.of(values.asOutput().shape().size())));
-    SetDiff1d<T, TInt32> diff = tf.setDiff1d(flatValues, allowedValues, TInt32.DTYPE);
+    SetDiff1d<T, TInt32> diff = tf.setDiff1d(flatValues, allowedValues, TInt32.class);
     long diffSize = diff.out().asOutput().shape().size();
 
     if (diffSize != Shape.UNKNOWN_SIZE) {
@@ -409,7 +408,7 @@ public static <T extends TNumber> Operand<T> valueCheck(
             tf.withSubScope("valueCheck")
                 .withControlDependencies(Collections.singletonList(assertThat));
         return ltf.identity(values);
-      } else if (!cond.asOutput().data().getBoolean())
+      } else if (!cond.asOutput().asTensor().getBoolean())
         throw new IllegalArgumentException(String.format("%s : values not in value set", prefix));
       else return values;
     }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaDelta.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaDelta.java
index 0adf5f58910..822eb490f22 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaDelta.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaDelta.java
@@ -141,10 +141,10 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    */
   private <T extends TType> void createAdaDeltaSlot(Output<T> v) {
     Operand<T> accumulatorInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), ACCUMULATOR, accumulatorInitializer);
     Operand<T> updateInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), ACCUMULATOR_UPDATE, updateInitializer);
   }
 
@@ -157,9 +157,9 @@ protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable
         variable,
         accumSlot,
         accumUpdateSlot,
-        tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()),
-        tf.dtypes.cast(tf.constant(rho), gradient.dataType()),
-        tf.dtypes.cast(tf.constant(epsilon), gradient.dataType()),
+        tf.dtypes.cast(tf.constant(learningRate), gradient.type()),
+        tf.dtypes.cast(tf.constant(rho), gradient.type()),
+        tf.dtypes.cast(tf.constant(epsilon), gradient.type()),
         gradient);
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGrad.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGrad.java
index 1a7f4675662..08f5f18a9cd 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGrad.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGrad.java
@@ -131,7 +131,7 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    */
   private <T extends TType> void createAdaGradSlot(Output<T> v) {
     Operand<T> initializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(initialAccumulatorValue), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(initialAccumulatorValue), v.type()));
     createSlot(v.asOutput(), ACCUMULATOR, initializer);
   }
 
@@ -140,7 +140,7 @@ private <T extends TType> void createAdaGradSlot(Output<T> v) {
   protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable) {
     Variable<T> slot = getSlot(variable, ACCUMULATOR).get();
     return tf.train.applyAdagrad(
-        variable, slot, tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()), gradient);
+        variable, slot, tf.dtypes.cast(tf.constant(learningRate), gradient.type()), gradient);
   }
 
   /** {@inheritDoc} */
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGradDA.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGradDA.java
index f76217fda85..df624e41c4e 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGradDA.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/AdaGradDA.java
@@ -187,7 +187,7 @@ protected void createSlots(List<Output<? extends TType>> variables) {
     for (Output<? extends TType> v : variables) {
       createAdaGradDASlot(v);
     }
-    globalStep = tf.withName("adagrad-da-global-step").variable(Shape.scalar(), TInt64.DTYPE);
+    globalStep = tf.withName("adagrad-da-global-step").variable(Shape.scalar(), TInt64.class);
     Assign<TInt64> globalStepInitializer = tf.assign(globalStep, tf.constant(0L));
     graph.addInitializer(globalStepInitializer);
   }
@@ -199,10 +199,10 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    * @param <T> the datatype of the variable.
    */
   private <T extends TType> void createAdaGradDASlot(Output<T> v) {
-    Operand<T> initializer = tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+    Operand<T> initializer = tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), ACCUMULATOR, initializer);
     Operand<T> sqInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(initialAccumulatorValue), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(initialAccumulatorValue), v.type()));
     createSlot(v.asOutput(), SQUARED_ACCUMULATOR, sqInitializer);
   }
 
@@ -216,9 +216,9 @@ protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable
         gradSlot,
         gradSquaredSlot,
         gradient,
-        tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()),
-        tf.dtypes.cast(tf.constant(l1Strength), gradient.dataType()),
-        tf.dtypes.cast(tf.constant(l2Strength), gradient.dataType()),
+        tf.dtypes.cast(tf.constant(learningRate), gradient.type()),
+        tf.dtypes.cast(tf.constant(l1Strength), gradient.type()),
+        tf.dtypes.cast(tf.constant(l2Strength), gradient.type()),
         globalStep);
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adam.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adam.java
index 8f620678781..72598d12543 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adam.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adam.java
@@ -189,10 +189,10 @@ protected void createSlots(List<Output<? extends TType>> variables) {
     for (Output<? extends TType> v : variables) {
       createAdamSlot(v.asOutput());
     }
-    betaOnePower = tf.withName("beta1_power").variable(Shape.scalar(), TFloat32.DTYPE);
+    betaOnePower = tf.withName("beta1_power").variable(Shape.scalar(), TFloat32.class);
     Assign<TFloat32> betaOnePowerInit = tf.assign(betaOnePower, tf.constant(betaOne));
     graph.addInitializer(betaOnePowerInit);
-    betaTwoPower = tf.withName("beta2_power").variable(Shape.scalar(), TFloat32.DTYPE);
+    betaTwoPower = tf.withName("beta2_power").variable(Shape.scalar(), TFloat32.class);
     Assign<TFloat32> betaTwoPowerInit = tf.assign(betaTwoPower, tf.constant(betaTwo));
     graph.addInitializer(betaTwoPowerInit);
   }
@@ -215,10 +215,10 @@ protected Optional<Op> prepare(String scopeName) {
    */
   private <T extends TType> void createAdamSlot(Output<T> v) {
     Operand<T> firstMomentInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), FIRST_MOMENT, firstMomentInitializer);
     Operand<T> secondMomentInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), SECOND_MOMENT, secondMomentInitializer);
   }
 
@@ -231,12 +231,12 @@ protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable
         variable,
         firstMomentSlot,
         secondMomentSlot,
-        tf.dtypes.cast(betaOnePower, gradient.dataType()),
-        tf.dtypes.cast(betaTwoPower, gradient.dataType()),
-        tf.dtypes.cast(learningRateConst, gradient.dataType()),
-        tf.dtypes.cast(betaOneConst, gradient.dataType()),
-        tf.dtypes.cast(betaTwoConst, gradient.dataType()),
-        tf.dtypes.cast(epsilonConst, gradient.dataType()),
+        tf.dtypes.cast(betaOnePower, gradient.type()),
+        tf.dtypes.cast(betaTwoPower, gradient.type()),
+        tf.dtypes.cast(learningRateConst, gradient.type()),
+        tf.dtypes.cast(betaOneConst, gradient.type()),
+        tf.dtypes.cast(betaTwoConst, gradient.type()),
+        tf.dtypes.cast(epsilonConst, gradient.type()),
         gradient);
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adamax.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adamax.java
index 335d83cedfa..cd95bb3bd07 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adamax.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Adamax.java
@@ -137,7 +137,7 @@ protected void createSlots(List<Output<? extends TType>> variables) {
     for (Output<? extends TType> v : variables) {
       createAdamaxSlot(v.asOutput());
     }
-    betaOnePower = tf.withName("beta1_power").variable(Shape.scalar(), TFloat32.DTYPE);
+    betaOnePower = tf.withName("beta1_power").variable(Shape.scalar(), TFloat32.class);
     Assign<TFloat32> betaOnePowerInit = tf.assign(betaOnePower, tf.constant(betaOne));
     ((Graph) tf.scope().env()).addInitializer(betaOnePowerInit);
   }
@@ -150,10 +150,10 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    */
   private <T extends TType> void createAdamaxSlot(Output<T> v) {
     Operand<T> firstMomentInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), FIRST_MOMENT, firstMomentInitializer);
     Operand<T> secondMomentInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), SECOND_MOMENT, secondMomentInitializer);
   }
 
@@ -167,11 +167,11 @@ protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable
         variable,
         firstMomentSlot,
         secondMomentSlot,
-        tf.dtypes.cast(betaOnePower, gradient.dataType()),
-        tf.dtypes.cast(learningRateConst, gradient.dataType()),
-        tf.dtypes.cast(betaOneConst, gradient.dataType()),
-        tf.dtypes.cast(betaTwoConst, gradient.dataType()),
-        tf.dtypes.cast(epsilonConst, gradient.dataType()),
+        tf.dtypes.cast(betaOnePower, gradient.type()),
+        tf.dtypes.cast(learningRateConst, gradient.type()),
+        tf.dtypes.cast(betaOneConst, gradient.type()),
+        tf.dtypes.cast(betaTwoConst, gradient.type()),
+        tf.dtypes.cast(epsilonConst, gradient.type()),
         gradient);
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Ftrl.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Ftrl.java
index 04c34a2535e..66314d2ffe0 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Ftrl.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Ftrl.java
@@ -230,10 +230,10 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    */
   private <T extends TType> void createFtrlSlot(Output<T> v) {
     Operand<T> initializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(initialAccumulatorValue), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(initialAccumulatorValue), v.type()));
     createSlot(v.asOutput(), ACCUMULATOR, initializer);
     Operand<T> linearInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), LINEAR_ACCUMULATOR, linearInitializer);
   }
 
@@ -248,12 +248,12 @@ protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable
         accumSlot, // accum
         linearSlot, // linear
         gradient, // gradient
-        tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()), // lr
-        tf.dtypes.cast(tf.constant(l1RegularizationStrength), gradient.dataType()), // l1
-        tf.dtypes.cast(tf.constant(l2RegularizationStrength), gradient.dataType()), // l2
+        tf.dtypes.cast(tf.constant(learningRate), gradient.type()), // lr
+        tf.dtypes.cast(tf.constant(l1RegularizationStrength), gradient.type()), // l1
+        tf.dtypes.cast(tf.constant(l2RegularizationStrength), gradient.type()), // l2
         tf.dtypes.cast(
-            tf.constant(l2ShrinkageRegularizationStrength), gradient.dataType()), // l2Shrinkage
-        tf.dtypes.cast(tf.constant(learningRatePower), gradient.dataType()), // lrPower
+            tf.constant(l2ShrinkageRegularizationStrength), gradient.type()), // l2Shrinkage
+        tf.dtypes.cast(tf.constant(learningRatePower), gradient.type()), // lrPower
         options);
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/GradientDescent.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/GradientDescent.java
index e307855e636..a373b2e5b55 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/GradientDescent.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/GradientDescent.java
@@ -66,7 +66,7 @@ public GradientDescent(Graph graph, String name, float learningRate) {
   @Override
   protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable) {
     return tf.train.applyGradientDescent(
-        variable, tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()), gradient);
+        variable, tf.dtypes.cast(tf.constant(learningRate), gradient.type()), gradient);
   }
 
   /** {@inheritDoc} */
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Momentum.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Momentum.java
index 111727d26fa..f6640409d60 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Momentum.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Momentum.java
@@ -125,7 +125,7 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    * @param <T> the data type of the variable
    */
   private <T extends TType> void createMomentumSlot(Output<T> v) {
-    Operand<T> initializer = tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+    Operand<T> initializer = tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), MOMENTUM, initializer);
   }
 
@@ -136,9 +136,9 @@ protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable
     return tf.train.applyMomentum(
         variable,
         slot,
-        tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()),
+        tf.dtypes.cast(tf.constant(learningRate), gradient.type()),
         gradient,
-        tf.dtypes.cast(tf.constant(momentum), gradient.dataType()),
+        tf.dtypes.cast(tf.constant(momentum), gradient.type()),
         ApplyMomentum.useNesterov(useNesterov));
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Nadam.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Nadam.java
index 48e5135c952..f9900a8ee78 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Nadam.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Nadam.java
@@ -1,6 +1,5 @@
 package org.tensorflow.framework.optimizers;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Graph;
 import org.tensorflow.Operand;
 import org.tensorflow.Output;
@@ -142,15 +141,15 @@ protected void createSlots(List<Output<? extends TType>> variables) {
     for (Output<? extends TType> v : variables) {
       createNadamSlot(v.asOutput());
     }
-    betaOnePower = tf.withName("beta1_power").variable(Shape.scalar(), TFloat32.DTYPE);
+    betaOnePower = tf.withName("beta1_power").variable(Shape.scalar(), TFloat32.class);
     Assign<TFloat32> betaOnePowerInit = tf.assign(betaOnePower, tf.constant(betaOne));
     ((Graph) tf.scope().env()).addInitializer(betaOnePowerInit);
 
-    betaTwoPower = tf.withName("beta2_power").variable(Shape.scalar(), TFloat32.DTYPE);
+    betaTwoPower = tf.withName("beta2_power").variable(Shape.scalar(), TFloat32.class);
     Assign<TFloat32> betaTwoPowerInit = tf.assign(betaTwoPower, tf.constant(betaTwo));
     ((Graph) tf.scope().env()).addInitializer(betaTwoPowerInit);
 
-    momentum = tf.withName("momentum").variable(Shape.scalar(), TFloat32.DTYPE);
+    momentum = tf.withName("momentum").variable(Shape.scalar(), TFloat32.class);
     Assign<TFloat32> momentumInit = tf.assign(momentum, tf.constant(1.0F));
     ((Graph) tf.scope().env()).addInitializer(momentumInit);
   }
@@ -163,14 +162,14 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    */
   private <T extends TType> void createNadamSlot(Output<T> v) {
     Operand<T> firstMomentInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), FIRST_MOMENT, firstMomentInitializer);
     Operand<T> secondMomentInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), SECOND_MOMENT, secondMomentInitializer);
 
     Operand<T> momentumInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(1.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(1.0f), v.type()));
     createSlot(v.asOutput(), MOMENTUM, momentumInitializer);
   }
 
@@ -198,7 +197,7 @@ protected Optional<Op> prepare(String scopeName) {
                     point5,
                     tf.math.pow(
                         decayBaseConst,
-                        tf.math.mul(decayConst, tf.dtypes.cast(localStepConst, TFloat32.DTYPE))))));
+                        tf.math.mul(decayConst, tf.dtypes.cast(localStepConst, TFloat32.class))))));
 
     mT1 =
         tf.math.mul(
@@ -209,7 +208,7 @@ protected Optional<Op> prepare(String scopeName) {
                     point5,
                     tf.math.pow(
                         decayBaseConst,
-                        tf.math.mul(decayConst, tf.dtypes.cast(nextStepConst, TFloat32.DTYPE))))));
+                        tf.math.mul(decayConst, tf.dtypes.cast(nextStepConst, TFloat32.class))))));
 
     Operand<TFloat32> mScheduleNew = tf.math.mul(momentum, mT);
 
@@ -222,57 +221,57 @@ protected Optional<Op> prepare(String scopeName) {
     oneMinusMScheduleNew = tf.math.sub(one, mScheduleNew);
     oneMinusMScheduleNext = tf.math.sub(one, mScheduleNext);
     vTPrimeDenominator =
-        tf.math.sub(one, tf.math.pow(betaTwoConst, tf.dtypes.cast(localStepConst, TFloat32.DTYPE)));
+        tf.math.sub(one, tf.math.pow(betaTwoConst, tf.dtypes.cast(localStepConst, TFloat32.class)));
     return Optional.empty();
   }
 
   /** {@inheritDoc} */
   @Override
   protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable) {
-    DataType<T> dType = gradient.dataType();
+    Class<T> type = gradient.type();
     Variable<T> m = getSlot(variable, FIRST_MOMENT).get(); // first Moment
     Variable<T> v = getSlot(variable, SECOND_MOMENT).get(); // Second Moment
 
     //  gPrime = grad / coefficients['oneMinusMScheduleNew']
-    Operand<T> gPrime = tf.math.div(gradient, tf.dtypes.cast(oneMinusMScheduleNew, dType));
+    Operand<T> gPrime = tf.math.div(gradient, tf.dtypes.cast(oneMinusMScheduleNew, type));
     // mT = (coefficients['beta_1_t'] * m + coefficients['one_minus_beta_1_t'] * grad)
     Operand<T> mT =
         tf.math.add(
-            tf.math.mul(tf.dtypes.cast(betaOneConst, dType), m),
-            tf.math.mul(tf.dtypes.cast(oneMinusBeta1, dType), gradient));
+            tf.math.mul(tf.dtypes.cast(betaOneConst, type), m),
+            tf.math.mul(tf.dtypes.cast(oneMinusBeta1, type), gradient));
     // mT = state_ops.assign(m, mT, use_locking=self._use_locking)
     // update m
     mT = tf.assign(m, mT, Assign.useLocking(true));
 
     // mTPrime = mT / coefficients['oneMinusMScheduleNext']
-    Operand<T> mTPrime = tf.math.div(mT, tf.dtypes.cast(oneMinusMScheduleNext, dType));
+    Operand<T> mTPrime = tf.math.div(mT, tf.dtypes.cast(oneMinusMScheduleNext, type));
 
     // vT = (coefficients['beta_2_t'] * v + coefficients['one_minus_beta_2_t'] *
     // math_ops.square(grad))
     Operand<T> vT =
         tf.math.add(
-            tf.math.mul(tf.dtypes.cast(betaTwoConst, dType), v),
-            tf.math.mul(tf.dtypes.cast(oneMinusBeta2, dType), tf.math.square(gradient)));
+            tf.math.mul(tf.dtypes.cast(betaTwoConst, type), v),
+            tf.math.mul(tf.dtypes.cast(oneMinusBeta2, type), tf.math.square(gradient)));
     // vT = state_ops.assign(v, vT, use_locking=self._use_locking)
     // update v
     vT = tf.assign(v, vT, Assign.useLocking(true));
 
     // vTPrime = vT / coefficients['vTPrimeDenominator']
-    Operand<T> vTPrime = tf.math.div(vT, tf.dtypes.cast(vTPrimeDenominator, dType));
+    Operand<T> vTPrime = tf.math.div(vT, tf.dtypes.cast(vTPrimeDenominator, type));
 
     // m_t_bar = (coefficients['oneMinusMT'] * gPrime + coefficients['mT1'] * mTPrime)
     Operand<T> m_t_bar =
         tf.math.add(
-            tf.math.mul(tf.dtypes.cast(oneMinusMT, dType), gPrime),
-            tf.math.mul(tf.dtypes.cast(mT1, dType), mTPrime));
+            tf.math.mul(tf.dtypes.cast(oneMinusMT, type), gPrime),
+            tf.math.mul(tf.dtypes.cast(mT1, type), mTPrime));
     // varT = var - coefficients['lr_t'] * m_t_bar / (math_ops.sqrt(vTPrime) +
     // coefficients['epsilon'])
     Operand<T> varT =
         tf.math.sub(
             variable,
             tf.math.div(
-                tf.math.mul(tf.dtypes.cast(learningRateConst, dType), m_t_bar),
-                tf.math.add(tf.math.sqrt(vTPrime), tf.dtypes.cast(epsilonConst, dType))));
+                tf.math.mul(tf.dtypes.cast(learningRateConst, type), m_t_bar),
+                tf.math.add(tf.math.sqrt(vTPrime), tf.dtypes.cast(epsilonConst, type))));
 
     return tf.assign(variable, varT, Assign.useLocking(true));
   }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Optimizer.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Optimizer.java
index 70f065814f7..fdf56da4a67 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Optimizer.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/Optimizer.java
@@ -220,7 +220,7 @@ private <T extends TType> Optional<Variable<T>> getSlot(String varName, String s
   protected <T extends TType> void createSlot(
       Output<T> variable, String slotName, Operand<T> initializer) {
     Variable<T> slot =
-        tf.withName(createName(variable, slotName)).variable(variable.shape(), variable.dataType());
+        tf.withName(createName(variable, slotName)).variable(variable.shape(), variable.type());
     Assign<T> slotInit = tf.assign(slot, initializer);
     graph.addInitializer(slotInit);
     String varName = variable.op().name();
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/RMSProp.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/RMSProp.java
index 9a48a9b8a7a..b3729dc367f 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/RMSProp.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/optimizers/RMSProp.java
@@ -175,14 +175,14 @@ protected void createSlots(List<Output<? extends TType>> variables) {
    */
   private <T extends TType> void createRMSPropSlot(Output<T> v) {
     Operand<T> rmsInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(1.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(1.0f), v.type()));
     createSlot(v.asOutput(), RMS, rmsInitializer);
     Operand<T> momentumInitializer =
-        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+        tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
     createSlot(v.asOutput(), MOMENTUM, momentumInitializer);
     if (centered) {
       Operand<T> mgInitializer =
-          tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.dataType()));
+          tf.fill(tf.shape(v), tf.dtypes.cast(tf.constant(0.0f), v.type()));
       createSlot(v.asOutput(), MG, mgInitializer);
     }
   }
@@ -199,20 +199,20 @@ protected <T extends TType> Op applyDense(Output<T> gradient, Output<T> variable
           mgSlot,
           rmsSlot,
           momentumSlot,
-          tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()),
-          tf.dtypes.cast(tf.constant(decay), gradient.dataType()),
-          tf.dtypes.cast(tf.constant(momentum), gradient.dataType()),
-          tf.dtypes.cast(tf.constant(epsilon), gradient.dataType()),
+          tf.dtypes.cast(tf.constant(learningRate), gradient.type()),
+          tf.dtypes.cast(tf.constant(decay), gradient.type()),
+          tf.dtypes.cast(tf.constant(momentum), gradient.type()),
+          tf.dtypes.cast(tf.constant(epsilon), gradient.type()),
           gradient);
     }
     return tf.train.applyRmsProp(
         variable,
         rmsSlot,
         momentumSlot,
-        tf.dtypes.cast(tf.constant(learningRate), gradient.dataType()),
-        tf.dtypes.cast(tf.constant(decay), gradient.dataType()),
-        tf.dtypes.cast(tf.constant(momentum), gradient.dataType()),
-        tf.dtypes.cast(tf.constant(epsilon), gradient.dataType()),
+        tf.dtypes.cast(tf.constant(learningRate), gradient.type()),
+        tf.dtypes.cast(tf.constant(decay), gradient.type()),
+        tf.dtypes.cast(tf.constant(momentum), gradient.type()),
+        tf.dtypes.cast(tf.constant(epsilon), gradient.type()),
         gradient);
   }
 
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/CastHelper.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/CastHelper.java
index aec75e6078a..4e310ff56e3 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/CastHelper.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/CastHelper.java
@@ -15,7 +15,6 @@
  */
 package org.tensorflow.framework.utils;
 
-import org.tensorflow.DataType;
 import org.tensorflow.Operand;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.family.TType;
@@ -35,8 +34,8 @@ public class CastHelper {
    */
   @SuppressWarnings("unchecked")
   public static <T extends TType, U extends TType> Operand<T> cast(
-          Ops tf, Operand<U> value, DataType<T> requiredType) {
-    return (value.asOutput().dataType() == requiredType)
+          Ops tf, Operand<U> value, Class<T> requiredType) {
+    return (value.asOutput().type() == requiredType)
         ? (Operand<T>) value
         : tf.dtypes.cast(value, requiredType);
   }
diff --git a/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/ShapeUtils.java b/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/ShapeUtils.java
index 122de9f21ae..743e8114f6e 100644
--- a/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/ShapeUtils.java
+++ b/tensorflow-framework/src/main/java/org/tensorflow/framework/utils/ShapeUtils.java
@@ -17,10 +17,14 @@
 import org.tensorflow.*;
 import org.tensorflow.ndarray.NdArray;
 import org.tensorflow.ndarray.Shape;
+import org.tensorflow.ndarray.buffer.DataBuffer;
+import org.tensorflow.ndarray.buffer.DataBuffers;
+import org.tensorflow.ndarray.buffer.IntDataBuffer;
 import org.tensorflow.op.Scope;
 import org.tensorflow.types.TInt32;
 import org.tensorflow.types.TInt64;
 import org.tensorflow.types.TUint8;
+import org.tensorflow.types.family.TInteger;
 import org.tensorflow.types.family.TNumber;
 
 import java.util.ArrayList;
@@ -36,7 +40,7 @@ public class ShapeUtils {
    * @param dims the Operand containing the shape values
    * @return a new Shape based on an Operand that contains dimensions
    */
-  public static <T extends TNumber> Shape toShape(Scope scope, Operand<T> dims) {
+  public static <T extends TInteger> Shape toShape(Scope scope, Operand<T> dims) {
     long[] longDims = getLongArray(scope, dims);
     return Shape.of(longDims);
   }
@@ -62,65 +66,37 @@ public static int[] getIntArray(Scope scope, Operand<TInt32> dims) {
    * @return the long array
    * @throws java.lang.IllegalArgumentException if the dims type is not an integer
    */
-  public static <T extends TNumber> long[] getLongArray(Scope scope, Operand<T> dims) {
-    DataType<T> dType = dims.asOutput().dataType();
-    if (!dType.isInteger()) {
-      throw new IllegalArgumentException("the data type must be an integer type");
-    }
-    List<Long> result = new ArrayList<>();
-
+  public static <T extends TInteger> long[] getLongArray(Scope scope, Operand<T> dims) {
     if (scope.env().isEager()) {
-      if (dType.equals(TInt32.DTYPE)) {
-        @SuppressWarnings("unchecked")
-        Operand<TInt32> idims = (Operand<TInt32>) dims;
-
-        idims.asOutput().data().scalars().forEach(s -> result.add((long) s.getInt()));
-      } else if (dType.equals(TInt64.DTYPE)) {
-        @SuppressWarnings("unchecked")
-        Operand<TInt64> ldims = (Operand<TInt64>) dims;
-        ldims.asOutput().data().scalars().forEach(s -> result.add(s.getLong()));
-      } else if (dType.equals(TUint8.DTYPE)) {
-        @SuppressWarnings("unchecked")
-        Operand<TUint8> udims = (Operand<TUint8>) dims;
-        udims.asOutput().data().scalars().forEach(s -> result.add(s.getObject().longValue()));
-      } else { // shouldn't happen
-        throw new IllegalArgumentException("the data type must be an integer type");
-      }
-
-    } else {
-      try (Session session = new Session((Graph) scope.env())) {
-        if (dType.equals(TInt32.DTYPE)) {
-          try (Tensor<TInt32> tensorResult =
-              session.runner().fetch(dims).run().get(0).expect(TInt32.DTYPE)) {
-            tensorResult.data().scalars().forEach(s -> result.add((long) s.getInt()));
-          }
-        } else if (dType.equals(TInt64.DTYPE)) {
-          try (Tensor<TInt64> tensorResult =
-              session.runner().fetch(dims).run().get(0).expect(TInt64.DTYPE)) {
-            tensorResult.data().scalars().forEach(s -> result.add(s.getLong()));
-          }
-        } else if (dType.equals(TUint8.DTYPE)) {
-          try (Tensor<TUint8> tensorResult =
-              session.runner().fetch(dims).run().get(0).expect(TUint8.DTYPE)) {
-            tensorResult.data().scalars().forEach(s -> result.add(s.getObject().longValue()));
-          }
-        } else { // shouldn't happen
-          throw new IllegalArgumentException("the data type must be an integer type");
-        }
-      }
+      return getLongArray(dims.asTensor());
+    }
+    try (Session session = new Session((Graph)scope.env());
+        T tensor = session.runner().fetch(dims).run().get(0)) {
+      return getLongArray(tensor);
     }
-    return result.stream().mapToLong(i -> i).toArray();
   }
 
   /**
-   * Gets the shape for the data within a Tensor
+   * Converts a TInt32 or TInt64 to a java long array
    *
-   * @param tensor the tensor
-   * @return the Shape of the tensor's data;
+   * @param scope the TensorFlow scope
+   * @param dims the dimension tensor
+   * @param <T> the type of the dimensions, must either be TInt32 or TInt64 type
+   * @return the long array
+   * @throws java.lang.IllegalArgumentException if the dims type is not an integer
    */
-  public static <T extends TNumber> Shape getShape(Tensor<T> tensor) {
-    NdArray<?> data = (NdArray<?>) tensor.data();
-    return data.shape();
+  public static <T extends TInteger> long[] getLongArray(T dims) {
+    List<Long> result = new ArrayList<>();
+    if (dims instanceof TInt32) {
+      ((TInt32)dims).scalars().forEach(s -> result.add((long) s.getInt()));
+    } else if (dims instanceof TInt64) {
+      ((TInt64)dims).scalars().forEach(s -> result.add(s.getLong()));
+    } else if (dims instanceof TUint8) {
+      ((TUint8)dims).scalars().forEach(s -> result.add(s.getObject().longValue()));
+    } else { // shouldn't happen
+      throw new IllegalArgumentException("the data type must be an integer type");
+    }
+    return result.stream().mapToLong(i -> i).toArray();
   }
 
   /**
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/activations/ReLUTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/activations/ReLUTest.java
index f54401515ab..a0aa2c4b453 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/activations/ReLUTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/activations/ReLUTest.java
@@ -90,8 +90,8 @@ public void testCallFloat16() {
         Ops tf = session.getTF();
         ReLU<TFloat16> instance = new ReLU<>(tf);
         Operand<TFloat16> result =
-            instance.call(tf.dtypes.cast(tf.constant(input), TFloat16.DTYPE));
-        session.evaluate(tf.dtypes.cast(tf.constant(expected), TFloat16.DTYPE), result);
+            instance.call(tf.dtypes.cast(tf.constant(input), TFloat16.class));
+        session.evaluate(tf.dtypes.cast(tf.constant(expected), TFloat16.class), result);
       }
   }
 
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/BatchDatasetTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/BatchDatasetTest.java
index 6a54cb08de6..41be23a8f6f 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/BatchDatasetTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/BatchDatasetTest.java
@@ -17,7 +17,6 @@
 
 import org.junit.jupiter.api.Test;
 import org.tensorflow.Operand;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt32;
 
@@ -40,19 +39,15 @@ public void testEagerBatchDataset() {
             Arrays.asList(
                 tf.constant(testMatrix1),
                 tf.constant(testMatrix2)),
-            Arrays.asList(TInt32.DTYPE, TInt32.DTYPE))
+            Arrays.asList(TInt32.class, TInt32.class))
         .batch(2);
 
     int count = 0;
     for (List<Operand<?>> components : dataset) {
-      try (Tensor<TInt32> batch1 =
-               components.get(0).asTensor().expect(TInt32.DTYPE);
-           Tensor<TInt32> batch2 =
-               components.get(1).asTensor().expect(TInt32.DTYPE);) {
-
-        assertEquals(testMatrix1.slice(range(count, count + 2)), batch1.data());
-        assertEquals(testMatrix2.slice(range(count, count + 2)), batch2.data());
-
+      try (TInt32 batch1 = (TInt32)components.get(0).asTensor();
+           TInt32 batch2 = (TInt32)components.get(1).asTensor()) {
+        assertEquals(testMatrix1.slice(range(count, count + 2)), batch1);
+        assertEquals(testMatrix2.slice(range(count, count + 2)), batch2);
         count += 2;
       }
     }
@@ -63,23 +58,16 @@ public void testDropLastBatch() {
     Ops tf = Ops.create();
     Dataset dataset = Dataset
         .fromTensorSlices(tf,
-            Arrays.asList(
-                tf.constant(testMatrix1),
-                tf.constant(testMatrix2)),
-            Arrays.asList(TInt32.DTYPE, TInt32.DTYPE))
+            Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2)),
+            Arrays.asList(TInt32.class, TInt32.class))
         .batch(3, true);
 
     int count = 0;
     for (List<Operand<?>> components : dataset) {
-
-      try (Tensor<TInt32> batch1 =
-               components.get(0).asTensor().expect(TInt32.DTYPE);
-           Tensor<TInt32> batch2 =
-               components.get(1).asTensor().expect(TInt32.DTYPE);) {
-
-        assertEquals(testMatrix1.slice(range(count, count + 3)), batch1.data());
-        assertEquals(testMatrix2.slice(range(count, count + 3)), batch2.data());
-
+      try (TInt32 batch1 = (TInt32)components.get(0).asTensor();
+          TInt32 batch2 = (TInt32)components.get(1).asTensor()) {
+        assertEquals(testMatrix1.slice(range(count, count + 3)), batch1);
+        assertEquals(testMatrix2.slice(range(count, count + 3)), batch2);
         count += 3;
       }
     }
@@ -90,31 +78,23 @@ public void testKeepLastBatch() {
     Ops tf = Ops.create();
     Dataset dataset = Dataset
         .fromTensorSlices(tf,
-            Arrays.asList(
-                tf.constant(testMatrix1),
-                tf.constant(testMatrix2)),
-            Arrays.asList(TInt32.DTYPE, TInt32.DTYPE))
+            Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2)),
+            Arrays.asList(TInt32.class, TInt32.class))
         .batch(3, false);
 
     int count = 0;
     boolean foundLastBatch = false;
 
     for (List<Operand<?>> components : dataset) {
-      try (Tensor<TInt32> batch1 =
-               components.get(0).asTensor().expect(TInt32.DTYPE);
-           Tensor<TInt32> batch2 =
-               components.get(1).asTensor().expect(TInt32.DTYPE);) {
+      try (TInt32 batch1 = (TInt32)components.get(0).asTensor();
+           TInt32 batch2 = (TInt32)components.get(1).asTensor();) {
         if (count == 0) {
-          assertEquals(testMatrix1.slice(range(count, count + 3)),
-              batch1.data());
-          assertEquals(testMatrix2.slice(range(count, count + 3)),
-              batch2.data());
+          assertEquals(testMatrix1.slice(range(count, count + 3)), batch1);
+          assertEquals(testMatrix2.slice(range(count, count + 3)), batch2);
           count += 3;
         } else {
-          assertEquals(testMatrix1.slice(range(count, count + 1)),
-              batch1.data());
-          assertEquals(testMatrix2.slice(range(count, count + 1)),
-              batch2.data());
+          assertEquals(testMatrix1.slice(range(count, count + 1)), batch1);
+          assertEquals(testMatrix2.slice(range(count, count + 1)), batch2);
           foundLastBatch = true;
         }
       }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/DatasetIteratorTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/DatasetIteratorTest.java
index 6bb6e21f330..c5cb3978b4b 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/DatasetIteratorTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/DatasetIteratorTest.java
@@ -16,17 +16,17 @@
 package org.tensorflow.framework.data;
 
 import org.junit.jupiter.api.Test;
-import org.tensorflow.DataType;
 import org.tensorflow.Graph;
 import org.tensorflow.Operand;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
+import org.tensorflow.types.family.TType;
 import org.tensorflow.exceptions.TFOutOfRangeException;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt32;
 
 import java.util.Arrays;
 import java.util.List;
+import org.tensorflow.util.TensorList;
 
 import static org.junit.jupiter.api.Assertions.assertEquals;
 
@@ -39,7 +39,7 @@ public void testGraphIteration() {
 
       List<Operand<?>> tensors = Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2));
 
-      List<DataType<?>> dataTypes = Arrays.asList(TInt32.DTYPE, TInt32.DTYPE);
+      List<Class<? extends TType>> dataTypes = Arrays.asList(TInt32.class, TInt32.class);
 
       Dataset dataset = Dataset.fromTensorSlices(tf, tensors, dataTypes);
       DatasetIterator iterator = dataset.makeOneShotIterator();
@@ -53,15 +53,12 @@ public void testGraphIteration() {
 
         int batches = 0;
         while (true) {
-          try {
-            List<Tensor<?>> outputs = session.runner().fetch(x).fetch(y).run();
-
-            try (Tensor<TInt32> xBatch = outputs.get(0).expect(TInt32.DTYPE);
-                Tensor<TInt32> yBatch = outputs.get(1).expect(TInt32.DTYPE)) {
-              assertEquals(testMatrix1.get(batches), xBatch.data());
-              assertEquals(testMatrix2.get(batches), yBatch.data());
-              batches++;
-            }
+          try (TensorList outputs = session.runner().fetch(x).fetch(y).run()) {
+            TInt32 xBatch = outputs.get(0);
+            TInt32 yBatch = outputs.get(1);
+            assertEquals(testMatrix1.get(batches), xBatch);
+            assertEquals(testMatrix2.get(batches), yBatch);
+            batches++;
           } catch (TFOutOfRangeException e) {
             break;
           }
@@ -77,17 +74,15 @@ public void testEagerIteration() {
 
     List<Operand<?>> tensors = Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2));
 
-    List<DataType<?>> dataTypes = Arrays.asList(TInt32.DTYPE, TInt32.DTYPE);
+    List<Class<? extends TType>> dataTypes = Arrays.asList(TInt32.class, TInt32.class);
 
     Dataset dataset = Dataset.fromTensorSlices(tf, tensors, dataTypes);
     int count = 0;
     for (List<Operand<?>> outputs : dataset) {
-      try (Tensor<TInt32> batch1 = outputs.get(0).asTensor().expect(TInt32.DTYPE);
-          Tensor<TInt32> batch2 = outputs.get(1).asTensor().expect(TInt32.DTYPE); ) {
-
-        assertEquals(testMatrix1.get(count), batch1.data());
-        assertEquals(testMatrix2.get(count), batch2.data());
-
+      try (TInt32 batch1 = (TInt32)outputs.get(0).asTensor();
+          TInt32 batch2 = (TInt32)outputs.get(1).asTensor()) {
+        assertEquals(testMatrix1.get(count), batch1);
+        assertEquals(testMatrix2.get(count), batch2);
         count++;
       }
     }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/MapDatasetTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/MapDatasetTest.java
index 5960442ff70..c6be6960089 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/MapDatasetTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/MapDatasetTest.java
@@ -17,11 +17,10 @@
 
 import org.junit.jupiter.api.BeforeEach;
 import org.junit.jupiter.api.Test;
-import org.tensorflow.DataType;
 import org.tensorflow.Graph;
 import org.tensorflow.Operand;
 import org.tensorflow.Session;
-import org.tensorflow.Tensor;
+import org.tensorflow.types.family.TType;
 import org.tensorflow.exceptions.TFOutOfRangeException;
 import org.tensorflow.op.Ops;
 import org.tensorflow.ndarray.IntNdArray;
@@ -30,6 +29,7 @@
 
 import java.util.Arrays;
 import java.util.List;
+import org.tensorflow.util.TensorList;
 
 import static org.junit.jupiter.api.Assertions.assertEquals;
 
@@ -60,13 +60,13 @@ public void testGraphIteration() {
 
       List<Operand<?>> tensors = Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2));
 
-      List<DataType<?>> dataTypes = Arrays.asList(TInt32.DTYPE, TInt32.DTYPE);
+      List<Class<? extends TType>> dataTypes = Arrays.asList(TInt32.class, TInt32.class);
 
       Dataset dataset =
           Dataset.fromTensorSlices(tf, tensors, dataTypes)
               .mapAllComponents(
                   component ->
-                      tf.math.mul(component.asOutput().expect(TInt32.DTYPE), tf.constant(2)));
+                      tf.math.mul(component.asOutput().expect(TInt32.class), tf.constant(2)));
 
       DatasetIterator iterator = dataset.makeOneShotIterator();
       List<Operand<?>> components = iterator.getNext();
@@ -78,17 +78,12 @@ public void testGraphIteration() {
 
         int batches = 0;
         while (true) {
-          try {
-            List<Tensor<?>> outputs = session.runner().fetch(X).fetch(y).run();
-
-            try (Tensor<TInt32> XBatch = outputs.get(0).expect(TInt32.DTYPE);
-                Tensor<TInt32> yBatch = outputs.get(1).expect(TInt32.DTYPE)) {
-
-              assertEquals(mapped1.get(batches), XBatch.data());
-              assertEquals(mapped2.get(batches), yBatch.data());
-
+          try (TensorList outputs = session.runner().fetch(X).fetch(y).run()) {
+              TInt32 xBatch = outputs.get(0);
+              TInt32 yBatch = outputs.get(1);
+              assertEquals(mapped1.get(batches), xBatch);
+              assertEquals(mapped2.get(batches), yBatch);
               batches++;
-            }
           } catch (TFOutOfRangeException e) {
             break;
           }
@@ -105,21 +100,18 @@ public void testEagerIteration() {
 
     List<Operand<?>> tensors = Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2));
 
-    List<DataType<?>> dataTypes = Arrays.asList(TInt32.DTYPE, TInt32.DTYPE);
+    List<Class<? extends TType>> dataTypes = Arrays.asList(TInt32.class, TInt32.class);
 
     Dataset dataset =
         Dataset.fromTensorSlices(tf, tensors, dataTypes)
-            .mapAllComponents(
-                op -> tf.math.mul(op.asOutput().expect(TInt32.DTYPE), tf.constant(2)));
+            .mapAllComponents(op -> tf.math.mul(op.expect(TInt32.class), tf.constant(2)));
 
     int count = 0;
     for (List<Operand<?>> outputs : dataset) {
-      try (Tensor<TInt32> XBatch = outputs.get(0).asTensor().expect(TInt32.DTYPE);
-          Tensor<TInt32> yBatch = outputs.get(1).asTensor().expect(TInt32.DTYPE); ) {
-
-        assertEquals(mapped1.get(count), XBatch.data());
-        assertEquals(mapped2.get(count), yBatch.data());
-
+      try (TInt32 XBatch = (TInt32)outputs.get(0).asTensor();
+          TInt32 yBatch = (TInt32)outputs.get(1).asTensor()) {
+        assertEquals(mapped1.get(count), XBatch);
+        assertEquals(mapped2.get(count), yBatch);
         count++;
       }
     }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/SkipDatasetTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/SkipDatasetTest.java
index 9ff8080034d..d0cdb4527a5 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/SkipDatasetTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/SkipDatasetTest.java
@@ -17,7 +17,6 @@
 
 import org.junit.jupiter.api.Test;
 import org.tensorflow.Operand;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt32;
 
@@ -35,16 +34,15 @@ public void testEagerSkipDataset() {
         Dataset.fromTensorSlices(
                 tf,
                 Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2)),
-                Arrays.asList(TInt32.DTYPE, TInt32.DTYPE))
+                Arrays.asList(TInt32.class, TInt32.class))
             .skip(2);
 
     int count = 2;
     for (List<Operand<?>> components : dataset) {
-      try (Tensor<TInt32> batch1 = components.get(0).asTensor().expect(TInt32.DTYPE);
-          Tensor<TInt32> batch2 =
-              components.get(1).asTensor().expect(TInt32.DTYPE); ) {
-        assertEquals(testMatrix1.get(count), batch1.data());
-        assertEquals(testMatrix2.get(count), batch2.data());
+      try (TInt32 batch1 = (TInt32)components.get(0).asTensor();
+          TInt32 batch2 = (TInt32)components.get(1).asTensor()) {
+        assertEquals(testMatrix1.get(count), batch1);
+        assertEquals(testMatrix2.get(count), batch2);
         count++;
       }
     }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/TakeDatasetTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/TakeDatasetTest.java
index 4419f4660db..79a2e79c72e 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/data/TakeDatasetTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/data/TakeDatasetTest.java
@@ -17,7 +17,6 @@
 
 import org.junit.jupiter.api.Test;
 import org.tensorflow.Operand;
-import org.tensorflow.Tensor;
 import org.tensorflow.op.Ops;
 import org.tensorflow.types.TInt32;
 
@@ -36,16 +35,15 @@ public void testEagerTakeDataset() {
         Dataset.fromTensorSlices(
                 tf,
                 Arrays.asList(tf.constant(testMatrix1), tf.constant(testMatrix2)),
-                Arrays.asList(TInt32.DTYPE, TInt32.DTYPE))
+                Arrays.asList(TInt32.class, TInt32.class))
             .take(4);
 
     int count = 0;
     for (List<Operand<?>> components : dataset) {
-      try (Tensor<TInt32> batch1 = components.get(0).asTensor().expect(TInt32.DTYPE);
-          Tensor<TInt32> batch2 = components.get(1).asTensor().expect(TInt32.DTYPE); ) {
-
-        assertEquals(testMatrix1.get(count), batch1.data());
-        assertEquals(testMatrix2.get(count), batch2.data());
+      try (TInt32 batch1 = (TInt32)components.get(0).asTensor();
+          TInt32 batch2 = (TInt32)components.get(1).asTensor()) {
+        assertEquals(testMatrix1.get(count), batch1);
+        assertEquals(testMatrix2.get(count), batch2);
         count++;
       }
     }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ConstantTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ConstantTest.java
index 46e4232d5ae..4e81e0620e6 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ConstantTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ConstantTest.java
@@ -52,7 +52,7 @@ public void testCallUInt() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Constant<TUint8> instance = new Constant<>(tf, 0xf);
-        Operand<TUint8> operand = instance.call(tf.constant(shape), TUint8.DTYPE);
+        Operand<TUint8> operand = instance.call(tf.constant(shape), TUint8.class);
         session.evaluate(expected, operand);
       }
   }
@@ -68,7 +68,7 @@ public void testCallInt() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Constant<TInt32> instance = new Constant<>(tf, 0xf);
-        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.DTYPE);
+        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -84,7 +84,7 @@ public void testCallLong() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Constant<TInt64> instance = new Constant<>(tf, 0xffL);
-        Operand<TInt64> operand = instance.call(tf.constant(shape), TInt64.DTYPE);
+        Operand<TInt64> operand = instance.call(tf.constant(shape), TInt64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -98,7 +98,7 @@ public void testCallFloat() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Constant<TFloat32> instance = new Constant<>(tf, 12.F);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -113,7 +113,7 @@ public void testCallDouble() {
         Shape shape = Shape.of(2, 2);
 
         Constant<TFloat64> instance = new Constant<>(tf, 11.);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -130,7 +130,7 @@ public void testCallString() {
               Shape shape = Shape.of(2, 2);
 
               Constant<TString> instance = new Constant<>(tf, 22);
-              instance.call(tf.constant(shape), TString.DTYPE);
+              instance.call(tf.constant(shape), TString.class);
               fail("IllegalArgumentException  should have been thrown for TString");
             }
           });
@@ -146,7 +146,7 @@ public void testCallBool() {
         Boolean[] expected = {true, true, true, true};
 
         Constant<TBool> instance = new Constant<>(tf, true);
-        Operand<TBool> operand = instance.call(tf.constant(shape), TBool.DTYPE);
+        Operand<TBool> operand = instance.call(tf.constant(shape), TBool.class);
         session.evaluate(expected, operand);
       }
   }
@@ -159,8 +159,8 @@ public void testReproducible() {
         Shape shape = Shape.of(2, 2);
 
         Constant<TFloat64> instance = new Constant<>(tf, 11.);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/GlorotTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/GlorotTest.java
index a68bf2a0a98..e9769806928 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/GlorotTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/GlorotTest.java
@@ -51,9 +51,9 @@ public void testCallNormalFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        Glorot<TFloat32, TFloat32> instance = new Glorot<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Glorot<TFloat32> instance = new Glorot<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
 
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -68,8 +68,8 @@ public void testCallNormalDouble() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        Glorot<TFloat64, TFloat64> instance = new Glorot<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Glorot<TFloat64> instance = new Glorot<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -82,8 +82,8 @@ public void testCallUniformFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        Glorot<TFloat32, TFloat32> instance = new Glorot<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Glorot<TFloat32> instance = new Glorot<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -97,8 +97,8 @@ public void testCallUniformDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        Glorot<TFloat64, TFloat64> instance = new Glorot<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Glorot<TFloat64> instance = new Glorot<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -109,9 +109,9 @@ public void testCallNormalReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        Glorot<TFloat64, TFloat64> instance = new Glorot<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Glorot<TFloat64> instance = new Glorot<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -122,9 +122,9 @@ public void testCallUniformReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        Glorot<TFloat64, TFloat64> instance = new Glorot<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Glorot<TFloat64> instance = new Glorot<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -135,10 +135,10 @@ public void testCallNORMALReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        Glorot<TFloat64, TFloat64> instance =
+        Glorot<TFloat64> instance =
             new Glorot<>(tf, Distribution.NORMAL, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/HeTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/HeTest.java
index 468759d347f..8953fa3005e 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/HeTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/HeTest.java
@@ -51,8 +51,8 @@ public void testCallNormalFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        He<TFloat32, TFloat32> instance = new He<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        He<TFloat32> instance = new He<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -66,8 +66,8 @@ public void testCallNormalDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        He<TFloat64, TFloat64> instance = new He<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        He<TFloat64> instance = new He<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -80,8 +80,8 @@ public void testCallFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        He<TFloat32, TFloat32> instance = new He<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        He<TFloat32> instance = new He<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -95,8 +95,8 @@ public void testCallDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        He<TFloat64, TFloat64> instance = new He<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        He<TFloat64> instance = new He<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -107,9 +107,9 @@ public void testCallNormalReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        He<TFloat64, TFloat64> instance = new He<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        He<TFloat64> instance = new He<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -120,9 +120,9 @@ public void testCallUniformReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        He<TFloat64, TFloat64> instance = new He<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        He<TFloat64> instance = new He<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -133,9 +133,9 @@ public void testCallNORMALReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        He<TFloat64, TFloat64> instance = new He<>(tf, Distribution.NORMAL, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        He<TFloat64> instance = new He<>(tf, Distribution.NORMAL, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/IdentityTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/IdentityTest.java
index adb6c0c118a..6eee5473937 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/IdentityTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/IdentityTest.java
@@ -45,23 +45,6 @@ public void setUp() {}
   @AfterEach
   public void tearDown() {}
 
-  /** Test of call method, of class Orthogonal. */
-  @Test
-  public void testCallInt() {
-    for (TestSession.Mode tfMode : tfModes)
-      assertThrows(
-          java.lang.IllegalArgumentException.class,
-          () -> {
-            try (TestSession session = TestSession.createTestSession(tfMode)) {
-              Ops tf = session.getTF();
-              Shape shape = Shape.of(10, 10);
-              Identity<TInt32> instance = new Identity<>(tf, 2.);
-              instance.call(tf.constant(shape), TInt32.DTYPE);
-              fail("Should have thrown IllegalArgumentException on Integer type");
-            }
-          });
-  }
-
   /** Test of call method, of class Constant. */
   @Test
   public void testCallFloat() {
@@ -82,7 +65,7 @@ public void testCallFloat() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(10, 10);
         Identity<TFloat32> instance = new Identity<>(tf, 2.);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -108,7 +91,7 @@ public void testCallDouble() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(10, 10);
         Identity<TFloat64> instance = new Identity<>(tf, 2.);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -121,8 +104,8 @@ public void testReproducible() {
         Shape shape = Shape.of(2, 2);
 
         Identity<TFloat64> instance = new Identity<>(tf, 2.);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/LeCunTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/LeCunTest.java
index 6033f9e12a5..336850a5549 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/LeCunTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/LeCunTest.java
@@ -51,8 +51,8 @@ public void testCallNormalFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        LeCun<TFloat32, TFloat32> instance = new LeCun<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        LeCun<TFloat32> instance = new LeCun<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -66,8 +66,8 @@ public void testCallNormalDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        LeCun<TFloat64, TFloat64> instance = new LeCun<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        LeCun<TFloat64> instance = new LeCun<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -80,8 +80,8 @@ public void testCallUniformFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        LeCun<TFloat32, TFloat32> instance = new LeCun<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        LeCun<TFloat32> instance = new LeCun<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -95,8 +95,8 @@ public void testCallUniformDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        LeCun<TFloat64, TFloat64> instance = new LeCun<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        LeCun<TFloat64> instance = new LeCun<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -107,9 +107,9 @@ public void testCallNormalReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        LeCun<TFloat64, TFloat64> instance = new LeCun<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        LeCun<TFloat64> instance = new LeCun<>(tf, Distribution.TRUNCATED_NORMAL, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -120,9 +120,9 @@ public void testCallUniformReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        LeCun<TFloat64, TFloat64> instance = new LeCun<>(tf, Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        LeCun<TFloat64> instance = new LeCun<>(tf, Distribution.UNIFORM, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -133,9 +133,9 @@ public void testCallNORMALReproducible() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        LeCun<TFloat64, TFloat64> instance = new LeCun<>(tf, Distribution.NORMAL, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        LeCun<TFloat64> instance = new LeCun<>(tf, Distribution.NORMAL, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OnesTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OnesTest.java
index bbd2ba3d384..053ba5dd7ff 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OnesTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OnesTest.java
@@ -52,7 +52,7 @@ public void testCallUInt() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Ones<TUint8> instance = new Ones<>(tf);
-        Operand<TUint8> operand = instance.call(tf.constant(shape), TUint8.DTYPE);
+        Operand<TUint8> operand = instance.call(tf.constant(shape), TUint8.class);
         session.evaluate(expected, operand);
       }
   }
@@ -66,7 +66,7 @@ public void testCallInt() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Ones<TInt32> instance = new Ones<>(tf);
-        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.DTYPE);
+        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -80,7 +80,7 @@ public void testCallLong() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Ones<TInt64> instance = new Ones<>(tf);
-        Operand<TInt64> operand = instance.call(tf.constant(shape), TInt64.DTYPE);
+        Operand<TInt64> operand = instance.call(tf.constant(shape), TInt64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -94,7 +94,7 @@ public void testCallFloat() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Ones<TFloat32> instance = new Ones<>(tf);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -109,7 +109,7 @@ public void testCallDouble() {
         Shape shape = Shape.of(2, 2);
 
         Ones<TFloat64> instance = new Ones<>(tf);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -126,7 +126,7 @@ public void testCallString() {
               Shape shape = Shape.of(2, 2);
 
               Ones<TString> instance = new Ones<>(tf);
-              instance.call(tf.constant(shape), TString.DTYPE);
+              instance.call(tf.constant(shape), TString.class);
               fail("IllegalArgumentException should have been thrown for TString");
             }
           });
@@ -141,7 +141,7 @@ public void testCallBool() {
         Shape shape = Shape.of(2, 2);
 
         Ones<TBool> instance = new Ones<>(tf);
-        Operand<TBool> operand = instance.call(tf.constant(shape), TBool.DTYPE);
+        Operand<TBool> operand = instance.call(tf.constant(shape), TBool.class);
         session.evaluate(expected, operand);
       }
   }
@@ -154,8 +154,8 @@ public void testReproducible() {
         Shape shape = Shape.of(2, 2);
 
         Ones<TFloat64> instance = new Ones<>(tf);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OrthogonalTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OrthogonalTest.java
index a4fff5fd19c..22b89d9177c 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OrthogonalTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/OrthogonalTest.java
@@ -47,23 +47,6 @@ public void setUp() {}
   @AfterEach
   public void tearDown() {}
 
-  /** Test of call method, of class Orthogonal. */
-  @Test
-  public void testCallInt() {
-    for (TestSession.Mode tfMode : tfModes)
-      assertThrows(
-          java.lang.IllegalArgumentException.class,
-          () -> {
-            try (TestSession session = TestSession.createTestSession(tfMode)) {
-              Ops tf = session.getTF();
-              Shape shape = Shape.of(10, 10);
-              Orthogonal<TInt32, TInt32> instance = new Orthogonal<>(tf, GAIN_VALUE, SEED);
-              instance.call(tf.constant(shape), TInt32.DTYPE);
-              fail("Should have thrown IllegalArgumentException on Integer type");
-            }
-          });
-  }
-
   /** Test of call method, of class Orthogonal. */
   @Test
   public void testCallFloat() {
@@ -173,8 +156,8 @@ public void testCallFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(10, 10);
-        Orthogonal<TFloat32, TFloat32> instance = new Orthogonal<>(tf, GAIN_VALUE, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Orthogonal<TFloat32> instance = new Orthogonal<>(tf, GAIN_VALUE, SEED);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -288,8 +271,8 @@ public void testCallDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(10, 10);
-        Orthogonal<TFloat64, TFloat64> instance = new Orthogonal<>(tf, GAIN_VALUE, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Orthogonal<TFloat64> instance = new Orthogonal<>(tf, GAIN_VALUE, SEED);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -301,9 +284,9 @@ public void testReproducible() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        Orthogonal<TFloat64, TFloat64> instance = new Orthogonal<>(tf, GAIN_VALUE, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Orthogonal<TFloat64> instance = new Orthogonal<>(tf, GAIN_VALUE, SEED);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomNormalTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomNormalTest.java
index 50aec670503..3b2b3bdb243 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomNormalTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomNormalTest.java
@@ -52,9 +52,9 @@ public void testCalltestSoftmaxFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        RandomNormal<TFloat32, TFloat32> instance =
+        RandomNormal<TFloat32> instance =
             new RandomNormal<>(tf, MEAN_VALUE, STDDEV_VALUE, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -68,9 +68,9 @@ public void testCalltestSoftmaxDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        RandomNormal<TFloat64, TFloat64> instance =
+        RandomNormal<TFloat64> instance =
             new RandomNormal<>(tf, MEAN_VALUE, STDDEV_VALUE, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -82,10 +82,10 @@ public void testReproducible() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        RandomNormal<TFloat64, TFloat64> instance =
+        RandomNormal<TFloat64> instance =
             new RandomNormal<>(tf, MEAN_VALUE, STDDEV_VALUE, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomUniformTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomUniformTest.java
index d3f9af74209..23e26083a9b 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomUniformTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/RandomUniformTest.java
@@ -53,9 +53,9 @@ public void testCallInt() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        RandomUniform<TInt32, TFloat64> instance =
+        RandomUniform<TInt32> instance =
             new RandomUniform<>(tf, MIN_VALUE, MAX_VALUE, SEED);
-        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.DTYPE);
+        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -68,9 +68,9 @@ public void testCallFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        RandomUniform<TFloat32, TFloat32> instance =
+        RandomUniform<TFloat32> instance =
             new RandomUniform<>(tf, MIN_VALUE, MAX_VALUE, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -84,9 +84,9 @@ public void testCallDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        RandomUniform<TFloat64, TFloat64> instance =
+        RandomUniform<TFloat64> instance =
             new RandomUniform<>(tf, MIN_VALUE, MAX_VALUE, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -98,10 +98,10 @@ public void testReproducible() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        RandomUniform<TFloat64, TFloat64> instance =
+        RandomUniform<TFloat64> instance =
             new RandomUniform<>(tf, MIN_VALUE, MAX_VALUE, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/TruncatedNormalTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/TruncatedNormalTest.java
index 0a551df2f38..96bf915e199 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/TruncatedNormalTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/TruncatedNormalTest.java
@@ -52,9 +52,9 @@ public void testCallFloat() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        TruncatedNormal<TFloat32, TFloat32> instance =
+        TruncatedNormal<TFloat32> instance =
             new TruncatedNormal<>(tf, MEAN_VALUE, STDDEV_VALUE, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -68,9 +68,9 @@ public void testCallDouble() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        TruncatedNormal<TFloat64, TFloat64> instance =
+        TruncatedNormal<TFloat64> instance =
             new TruncatedNormal<>(tf, MEAN_VALUE, STDDEV_VALUE, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -82,10 +82,10 @@ public void testReproducible() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        TruncatedNormal<TFloat64, TFloat64> instance =
+        TruncatedNormal<TFloat64> instance =
             new TruncatedNormal<>(tf, MEAN_VALUE, STDDEV_VALUE, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/VarianceScalingTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/VarianceScalingTest.java
index 77e0dd7afc7..159affb07e2 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/VarianceScalingTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/VarianceScalingTest.java
@@ -50,14 +50,14 @@ public void testCallFloat1FanInTruncatedNormal() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        VarianceScaling<TFloat32, TFloat32> instance =
+        VarianceScaling<TFloat32> instance =
             new VarianceScaling<>(
                 tf,
                 1.0,
                 VarianceScaling.Mode.FAN_IN,
                 VarianceScaling.Distribution.TRUNCATED_NORMAL,
                 SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -71,14 +71,14 @@ public void testCallDouble1FanInTruncatedNormal() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        VarianceScaling<TFloat64, TFloat64> instance =
+        VarianceScaling<TFloat64> instance =
             new VarianceScaling<>(
                 tf,
                 1.0,
                 VarianceScaling.Mode.FAN_IN,
                 VarianceScaling.Distribution.TRUNCATED_NORMAL,
                 SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -91,14 +91,14 @@ public void testCallFloat1FanInNormal() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        VarianceScaling<TFloat32, TFloat32> instance =
+        VarianceScaling<TFloat32> instance =
             new VarianceScaling<>(
                 tf,
                 1.0,
                 VarianceScaling.Mode.FAN_IN,
                 VarianceScaling.Distribution.NORMAL,
                 SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -112,14 +112,14 @@ public void testCalltestSoftmaxDouble1FanInNormal() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        VarianceScaling<TFloat64, TFloat64> instance =
+        VarianceScaling<TFloat64> instance =
             new VarianceScaling<>(
                 tf,
                 1.0,
                 VarianceScaling.Mode.FAN_IN,
                 VarianceScaling.Distribution.NORMAL,
                 SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -132,10 +132,10 @@ public void testCalltestSoftmaxFloat1FanInUNIFORM() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        VarianceScaling<TFloat32, TFloat32> instance =
+        VarianceScaling<TFloat32> instance =
             new VarianceScaling<>(
                 tf, 1.0, VarianceScaling.Mode.FAN_IN, VarianceScaling.Distribution.UNIFORM, SEED);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -149,10 +149,10 @@ public void testCalltestSoftmaxDouble1FanInUNIFORM() {
       try (TestSession session = TestSession.createTestSession(tfMode)) {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
-        VarianceScaling<TFloat64, TFloat64> instance =
+        VarianceScaling<TFloat64> instance =
             new VarianceScaling<>(
                 tf, 1.0, VarianceScaling.Mode.FAN_IN, VarianceScaling.Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -164,11 +164,11 @@ public void testReproducible1() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        VarianceScaling<TFloat64, TFloat64> instance =
+        VarianceScaling<TFloat64> instance =
             new VarianceScaling<>(
                 tf, 1.0, VarianceScaling.Mode.FAN_IN, VarianceScaling.Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -180,15 +180,15 @@ public void testReproducible2() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        VarianceScaling<TFloat64, TFloat64> instance =
+        VarianceScaling<TFloat64> instance =
             new VarianceScaling<>(
                 tf,
                 1.0,
                 VarianceScaling.Mode.FAN_IN,
                 VarianceScaling.Distribution.NORMAL,
                 SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -200,15 +200,15 @@ public void testReproducible3() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        VarianceScaling<TFloat64, TFloat64> instance =
+        VarianceScaling<TFloat64> instance =
             new VarianceScaling<>(
                 tf,
                 1.0,
                 VarianceScaling.Mode.FAN_OUT,
                 VarianceScaling.Distribution.TRUNCATED_NORMAL,
                 SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
@@ -220,11 +220,11 @@ public void testReproducible4() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
 
-        VarianceScaling<TFloat64, TFloat64> instance =
+        VarianceScaling<TFloat64> instance =
             new VarianceScaling<>(
                 tf, 1.0, VarianceScaling.Mode.FAN_AVG, VarianceScaling.Distribution.UNIFORM, SEED);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ZerosTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ZerosTest.java
index 975678add19..21bad6ff360 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ZerosTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/initializers/ZerosTest.java
@@ -49,7 +49,7 @@ public void testCallUInt() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Zeros<TUint8> instance = new Zeros<>(tf);
-        Operand<TUint8> operand = instance.call(tf.constant(shape), TUint8.DTYPE);
+        Operand<TUint8> operand = instance.call(tf.constant(shape), TUint8.class);
         session.evaluate(expected, operand);
       }
   }
@@ -63,7 +63,7 @@ public void testCallInt() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Zeros<TInt32> instance = new Zeros<>(tf);
-        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.DTYPE);
+        Operand<TInt32> operand = instance.call(tf.constant(shape), TInt32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -77,7 +77,7 @@ public void testCallLong() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Zeros<TInt64> instance = new Zeros<>(tf);
-        Operand<TInt64> operand = instance.call(tf.constant(shape), TInt64.DTYPE);
+        Operand<TInt64> operand = instance.call(tf.constant(shape), TInt64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -91,7 +91,7 @@ public void testCallFloat() {
         Ops tf = session.getTF();
         Shape shape = Shape.of(2, 2);
         Zeros<TFloat32> instance = new Zeros<>(tf);
-        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.DTYPE);
+        Operand<TFloat32> operand = instance.call(tf.constant(shape), TFloat32.class);
         session.evaluate(expected, operand);
       }
   }
@@ -106,7 +106,7 @@ public void testCallDouble() {
         Shape shape = Shape.of(2, 2);
 
         Zeros<TFloat64> instance = new Zeros<>(tf);
-        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(expected, operand);
       }
   }
@@ -120,7 +120,7 @@ public void testCallString() {
         Shape shape = Shape.of(2, 2);
 
         Zeros<TString> instance = new Zeros<>(tf);
-        Operand<TString> operand = instance.call(tf.constant(shape), TString.DTYPE);
+        Operand<TString> operand = instance.call(tf.constant(shape), TString.class);
         session.evaluateString(operand, String::isEmpty);
       }
   }
@@ -135,7 +135,7 @@ public void testCallBool() {
         Shape shape = Shape.of(2, 2);
 
         Zeros<TBool> instance = new Zeros<>(tf);
-        Operand<TBool> operand = instance.call(tf.constant(shape), TBool.DTYPE);
+        Operand<TBool> operand = instance.call(tf.constant(shape), TBool.class);
         session.evaluate(expected, operand);
       }
   }
@@ -148,8 +148,8 @@ public void testReproducible() {
         Shape shape = Shape.of(2, 2);
 
         Zeros<TFloat64> instance = new Zeros<>(tf);
-        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.DTYPE);
-        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.DTYPE);
+        Operand<TFloat64> operand1 = instance.call(tf.constant(shape), TFloat64.class);
+        Operand<TFloat64> operand2 = instance.call(tf.constant(shape), TFloat64.class);
         session.evaluate(operand1, operand2);
       }
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaDeltaTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaDeltaTest.java
index 5c4ce542c65..bf5e2b1450b 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaDeltaTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaDeltaTest.java
@@ -89,8 +89,8 @@ public void testBasic() {
           float[] var1Init = {3.0F, 4.0F};
           float[] fgrads = {grad, grad};
           Shape shape = Shape.of(var0Init.length);
-          Variable<TFloat32> var0 = tf.withName("var0").variable(shape, TFloat32.DTYPE);
-          Variable<TFloat32> var1 = tf.withName("var1").variable(shape, TFloat32.DTYPE);
+          Variable<TFloat32> var0 = tf.withName("var0").variable(shape, TFloat32.class);
+          Variable<TFloat32> var1 = tf.withName("var1").variable(shape, TFloat32.class);
 
           Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
           Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradDATest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradDATest.java
index ef9053ff1eb..d5b2657a4fc 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradDATest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradDATest.java
@@ -64,8 +64,8 @@ public void testBasic() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradTest.java
index c5ae178b84c..d2eb63056e8 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdaGradTest.java
@@ -79,8 +79,8 @@ public void testBasic() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamTest.java
index 461fa75397f..4f2afa07972 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamTest.java
@@ -16,7 +16,6 @@
 
 import org.junit.jupiter.api.*;
 import org.tensorflow.Graph;
-import org.tensorflow.Tensor;
 import org.tensorflow.framework.utils.ND;
 import org.tensorflow.framework.utils.TestSession;
 import org.tensorflow.ndarray.FloatNdArray;
@@ -80,8 +79,8 @@ public void testBasic() {
       Ops tf = instance.getTF();
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
@@ -140,25 +139,23 @@ public void testBasic() {
           (float) Math.pow(beta1, step + 1), (float) Math.pow(beta2, step + 1)
         };
 
-        try (Tensor<TFloat32> result =
+        try (TFloat32 result =
             session
                 .getGraphSession()
                 .runner()
                 .fetch("beta1_power")
                 .run()
-                .get(0)
-                .expect(TFloat32.DTYPE)) {
-          result.data().scalars().forEach(f -> assertEquals(powers[0], f.getFloat(), epsilon1));
+                .get(0)) {
+          result.scalars().forEach(f -> assertEquals(powers[0], f.getFloat(), epsilon1));
         }
-        try (Tensor<TFloat32> result =
+        try (TFloat32 result =
             session
                 .getGraphSession()
                 .runner()
                 .fetch("beta2_power")
                 .run()
-                .get(0)
-                .expect(TFloat32.DTYPE)) {
-          result.data().scalars().forEach(f -> assertEquals(powers[1], f.getFloat(), epsilon1));
+                .get(0)) {
+          result.scalars().forEach(f -> assertEquals(powers[1], f.getFloat(), epsilon1));
         }
         session.run(update);
 
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamaxTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamaxTest.java
index de17395f76a..ea0e96c7fad 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamaxTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/AdamaxTest.java
@@ -16,7 +16,6 @@
 
 import org.junit.jupiter.api.*;
 import org.tensorflow.Graph;
-import org.tensorflow.Tensor;
 import org.tensorflow.framework.utils.ND;
 import org.tensorflow.framework.utils.TestSession;
 import org.tensorflow.ndarray.FloatNdArray;
@@ -101,8 +100,8 @@ public void testBasic() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
@@ -149,15 +148,14 @@ public void testBasic() {
         // Test powers
         final float beta1Power = (float) Math.pow(BETA_ONE_DEFAULT, step + 1);
 
-        try (Tensor<TFloat32> result =
+        try (TFloat32 result =
             session
                 .getGraphSession()
                 .runner()
                 .fetch("beta1_power")
                 .run()
-                .get(0)
-                .expect(TFloat32.DTYPE)) {
-          result.data().scalars().forEach(f -> assertEquals(beta1Power, f.getFloat(), epsilon1));
+                .get(0)) {
+          result.scalars().forEach(f -> assertEquals(beta1Power, f.getFloat(), epsilon1));
         }
         session.run(update);
 
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/FtrlTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/FtrlTest.java
index 597f8e52bcd..7698d76f957 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/FtrlTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/FtrlTest.java
@@ -76,8 +76,8 @@ public void testFtrlWithL1L2L2Shrinkage() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
@@ -141,8 +141,8 @@ public void testFtrlWithL1() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
@@ -207,8 +207,8 @@ public void testFtrlWithL1L2() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
@@ -273,8 +273,8 @@ public void doTestFtrlwithoutRegularization() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/GradientDescentTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/GradientDescentTest.java
index 4362c54d815..aefcc537979 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/GradientDescentTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/GradientDescentTest.java
@@ -61,8 +61,8 @@ public void testBasic() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/MomentumTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/MomentumTest.java
index bcfff97773d..ca72c4f415c 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/MomentumTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/MomentumTest.java
@@ -77,8 +77,8 @@ public void testBasic() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
@@ -130,8 +130,8 @@ public void testMomentum() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/NadamTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/NadamTest.java
index a583d74246b..4df0994dffc 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/NadamTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/NadamTest.java
@@ -16,7 +16,6 @@
 
 import org.junit.jupiter.api.*;
 import org.tensorflow.Graph;
-import org.tensorflow.Tensor;
 import org.tensorflow.framework.utils.ND;
 import org.tensorflow.framework.utils.TestSession;
 import org.tensorflow.ndarray.FloatNdArray;
@@ -102,8 +101,8 @@ public void testBasic() {
 
       Shape shape0 = Shape.of(var0Init.length);
       Shape shape1 = Shape.of(var1Init.length);
-      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+      Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+      Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
       Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
       Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
@@ -147,15 +146,14 @@ public void testBasic() {
       session.evaluate(var0Init, var0);
       session.evaluate(var1Init, var1);
 
-      try (Tensor<TFloat32> result =
+      try (TFloat32 result =
           session
               .getGraphSession()
               .runner()
               .fetch("momentum")
               .run()
-              .get(0)
-              .expect(TFloat32.DTYPE)) {
-        result.data().scalars().forEach(f -> assertEquals(1F, f.getFloat(), epsilon1));
+              .get(0)) {
+        result.scalars().forEach(f -> assertEquals(1F, f.getFloat(), epsilon1));
       }
       momentum = 1F;
 
@@ -167,15 +165,14 @@ public void testBasic() {
             Nadam.BETA_ONE_DEFAULT * (1F - 0.5F * (float) Math.pow(0.96F, (0.004F * (step + 1))));
         momentum = momentum * mut;
 
-        try (Tensor<TFloat32> result =
+        try (TFloat32 result =
             session
                 .getGraphSession()
                 .runner()
                 .fetch("momentum")
                 .run()
-                .get(0)
-                .expect(TFloat32.DTYPE)) {
-          result.data().scalars().forEach(f -> assertEquals(momentum, f.getFloat(), epsilon1));
+                .get(0)) {
+          result.scalars().forEach(f -> assertEquals(momentum, f.getFloat(), epsilon1));
         }
         mcache = ND.mul(mcache, momentum);
         FloatNdArray[] resultsNP = nadamUpdateNdArray(var0Np, grads0Np, step, m0, v0, mcache);
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/RMSPropTest.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/RMSPropTest.java
index 202fb21ef68..3b002cd1dbe 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/RMSPropTest.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/optimizers/RMSPropTest.java
@@ -87,8 +87,8 @@ public void testDense() {
 
         Shape shape0 = Shape.of(var0Init.length);
         Shape shape1 = Shape.of(var1Init.length);
-        Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.DTYPE);
-        Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.DTYPE);
+        Variable<TFloat32> var0 = tf.withName("var0").variable(shape0, TFloat32.class);
+        Variable<TFloat32> var1 = tf.withName("var1").variable(shape1, TFloat32.class);
 
         Assign<TFloat32> var0Initializer = tf.assign(var0, tf.constant(var0Init));
         Assign<TFloat32> var1Initializer = tf.assign(var1, tf.constant(var1Init));
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/EagerTestSession.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/EagerTestSession.java
index bca90211e50..f9033c49cc8 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/EagerTestSession.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/EagerTestSession.java
@@ -84,57 +84,57 @@ public EagerSession getEagerSession() {
   /** {@inheritDoc} */
   @Override
   public <T extends TNumber> void evaluate(double expected, Operand<T> input) {
-    DataType<T> dtype = input.asOutput().dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> type = input.asOutput().type();
+    if (type == TFloat32.class) {
       Operand<TFloat32> o = (Operand<TFloat32>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
       }
       index.set(0);
-      o.data().scalars().forEach(f -> assertEquals(expected, f.getFloat(), epsilon));
-    } else if (dtype == TFloat64.DTYPE) {
+      o.asTensor().scalars().forEach(f -> assertEquals(expected, f.getFloat(), epsilon));
+    } else if (type == TFloat64.class) {
       Operand<TFloat64> o = (Operand<TFloat64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
       }
       index.set(0);
-      o.data().scalars().forEach(f -> assertEquals(expected, f.getDouble(), epsilon));
-    } else if (dtype == TInt32.DTYPE) {
+      o.asTensor().scalars().forEach(f -> assertEquals(expected, f.getDouble(), epsilon));
+    } else if (type == TInt32.class) {
       Operand<TInt32> o = (Operand<TInt32>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
       }
       index.set(0);
-      o.data().scalars().forEach(f -> assertEquals((int) expected, f.getInt()));
-    } else if (dtype == TInt64.DTYPE) {
+      o.asTensor().scalars().forEach(f -> assertEquals((int) expected, f.getInt()));
+    } else if (type == TInt64.class) {
       Operand<TInt64> o = (Operand<TInt64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
       }
       index.set(0);
-      o.data().scalars().forEach(f -> assertEquals((long) expected, f.getLong()));
-    } else if (dtype == TUint8.DTYPE) {
+      o.asTensor().scalars().forEach(f -> assertEquals((long) expected, f.getLong()));
+    } else if (type == TUint8.class) {
       Operand<TUint8> o = (Operand<TUint8>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %x\n", index.getAndIncrement(), f.getByte()));
       }
       index.set(0);
-      o.data().scalars().forEach(f -> assertEquals((long) expected, f.getByte()));
+      o.asTensor().scalars().forEach(f -> assertEquals((long) expected, f.getByte()));
     }
   }
 
@@ -146,71 +146,71 @@ public <T extends TNumber> void evaluate(Number[] expected, Output<T> input) {
         expected.length,
         size,
         () -> String.format("expected length (%d) != to input length (%d)", expected.length, size));
-    DataType<T> dtype = input.dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> type = input.type();
+    if (type == TFloat32.class) {
       Output<TFloat32> o = (Output<TFloat32>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(
               f ->
                   assertEquals(
                       expected[index.getAndIncrement()].floatValue(), f.getFloat(), epsilon));
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (type == TFloat64.class) {
       Output<TFloat64> o = (Output<TFloat64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(
               f ->
                   assertEquals(
                       expected[index.getAndIncrement()].doubleValue(), f.getDouble(), epsilon));
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (type == TInt32.class) {
       Output<TInt32> o = (Output<TInt32>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> assertEquals(expected[index.getAndIncrement()].intValue(), f.getInt()));
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (type == TInt64.class) {
       Output<TInt64> o = (Output<TInt64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> assertEquals(expected[index.getAndIncrement()].longValue(), f.getLong()));
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (type == TUint8.class) {
       Output<TUint8> o = (Output<TUint8>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%x). %d\n", index.getAndIncrement(), f.getByte()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> assertEquals(expected[index.getAndIncrement()].byteValue(), f.getByte()));
     }
@@ -219,69 +219,69 @@ public <T extends TNumber> void evaluate(Number[] expected, Output<T> input) {
   /** {@inheritDoc} */
   @Override
   public <T extends TType> void evaluate(FloatNdArray expected, Output<T> input) {
-    DataType<T> dtype = input.dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> type = input.type();
+    if (type == TFloat32.class) {
       Output<TFloat32> o = (Output<TFloat32>) input;
       AtomicLong index = new AtomicLong();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(
               f -> assertEquals(expected.getFloat(index.getAndIncrement()), f.getFloat(), epsilon));
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (type == TFloat64.class) {
       Output<TFloat64> o = (Output<TFloat64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(
               f ->
                   assertEquals(expected.getFloat(index.getAndIncrement()), f.getDouble(), epsilon));
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (type == TInt32.class) {
       Output<TInt32> o = (Output<TInt32>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
       }
       index.set(0);
-      for (IntNdArray f : o.data().scalars()) {
+      for (IntNdArray f : o.asTensor().scalars()) {
         assertEquals((int) expected.getFloat(index.getAndIncrement()), f.getInt());
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (type == TInt64.class) {
       Output<TInt64> o = (Output<TInt64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(
               f -> assertEquals((long) expected.getFloat(index.getAndIncrement()), f.getLong()));
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (type == TUint8.class) {
       Output<TUint8> o = (Output<TUint8>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEach(f -> System.out.printf("%d). %x\n", index.getAndIncrement(), f.getByte()));
       }
       index.set(0);
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(
               f -> assertEquals((long) expected.getFloat(index.getAndIncrement()), f.getByte()));
@@ -296,10 +296,10 @@ public void evaluateString(Output<TString> input, Predicate<String> predicate) {
     if (debug) {
       if (isScalar) {
         System.out.printf(
-            "0). %b <==> %s\n", predicate.test(input.data().getObject()), input.data().getObject());
+            "0). %b <==> %s\n", predicate.test(input.asTensor().getObject()), input.asTensor().getObject());
       } else {
         input
-            .data()
+            .asTensor()
             .scalars()
             .forEachIndexed(
                 (idx, s) ->
@@ -310,9 +310,9 @@ public void evaluateString(Output<TString> input, Predicate<String> predicate) {
     }
     index.set(0);
     if (isScalar) {
-      assertTrue(predicate.test(input.data().getObject()));
+      assertTrue(predicate.test(input.asTensor().getObject()));
     } else {
-      input.data().scalars().forEachIndexed((idx, s) -> assertTrue(predicate.test(s.getObject())));
+      input.asTensor().scalars().forEachIndexed((idx, s) -> assertTrue(predicate.test(s.getObject())));
     }
   }
 
@@ -320,16 +320,16 @@ public void evaluateString(Output<TString> input, Predicate<String> predicate) {
   @Override
   public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predicate) {
     AtomicInteger index = new AtomicInteger();
-    DataType<T> dtype = input.asOutput().dataType();
+    Class<T> type = input.asOutput().type();
     boolean isScalar = input.shape().equals(Shape.scalar());
-    if (dtype == TFloat32.DTYPE) {
+    if (type == TFloat32.class) {
       Output<TFloat32> o = (Output<TFloat32>) input;
       if (debug) {
         if (isScalar) {
           System.out.printf(
-              "0). %b <==> %f\n", predicate.test(o.data().getFloat()), o.data().getFloat());
+              "0). %b <==> %f\n", predicate.test(o.asTensor().getFloat()), o.asTensor().getFloat());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
@@ -340,20 +340,20 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
       }
       index.set(0);
       if (isScalar) {
-        assertTrue(predicate.test(o.data().getFloat()));
+        assertTrue(predicate.test(o.asTensor().getFloat()));
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.data().getFloat())));
+            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.asTensor().getFloat())));
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (type == TFloat64.class) {
       Output<TFloat64> o = (Output<TFloat64>) input;
       if (debug) {
         if (isScalar) {
           System.out.printf(
-              "0). %b <==> %f\n", predicate.test(o.data().getDouble()), o.data().getDouble());
+              "0). %b <==> %f\n", predicate.test(o.asTensor().getDouble()), o.asTensor().getDouble());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
@@ -364,20 +364,20 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
       }
       index.set(0);
       if (isScalar) {
-        assertTrue(predicate.test(o.data().getDouble()));
+        assertTrue(predicate.test(o.asTensor().getDouble()));
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.data().getDouble())));
+            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.asTensor().getDouble())));
       }
-    } else if (dtype == TFloat16.DTYPE) {
+    } else if (type == TFloat16.class) {
       Output<TFloat16> o = (Output<TFloat16>) input;
       if (debug) {
         if (isScalar) {
           System.out.printf(
-              "0). %b <==> %f\n", predicate.test(o.data().getFloat()), o.data().getFloat());
+              "0). %b <==> %f\n", predicate.test(o.asTensor().getFloat()), o.asTensor().getFloat());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
@@ -388,20 +388,20 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
       }
       index.set(0);
       if (isScalar) {
-        assertTrue(predicate.test(o.data().getFloat()));
+        assertTrue(predicate.test(o.asTensor().getFloat()));
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.data().getFloat())));
+            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.asTensor().getFloat())));
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (type == TInt32.class) {
       Output<TInt32> o = (Output<TInt32>) input;
       if (debug) {
         if (isScalar) {
           System.out.printf(
-              "0). %b <==> %d\n", predicate.test(o.data().getInt()), o.data().getInt());
+              "0). %b <==> %d\n", predicate.test(o.asTensor().getInt()), o.asTensor().getInt());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
@@ -412,20 +412,20 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
       }
       index.set(0);
       if (isScalar) {
-        assertTrue(predicate.test(o.data().getInt()));
+        assertTrue(predicate.test(o.asTensor().getInt()));
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.data().getInt())));
+            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.asTensor().getInt())));
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (type == TInt64.class) {
       Output<TInt64> o = (Output<TInt64>) input;
       if (debug) {
         if (isScalar) {
           System.out.printf(
-              "0). %b <==> %d\n", predicate.test(o.data().getLong()), o.data().getLong());
+              "0). %b <==> %d\n", predicate.test(o.asTensor().getLong()), o.asTensor().getLong());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
@@ -436,20 +436,20 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
       }
       index.set(0);
       if (isScalar) {
-        assertTrue(predicate.test(o.data().getLong()));
+        assertTrue(predicate.test(o.asTensor().getLong()));
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.data().getLong())));
+            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.asTensor().getLong())));
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (type == TUint8.class) {
       Output<TUint8> o = (Output<TUint8>) input;
       if (debug) {
         if (isScalar) {
           System.out.printf(
-              "0). %b <==> %x\n", predicate.test(o.data().getByte()), o.data().getByte());
+              "0). %b <==> %x\n", predicate.test(o.asTensor().getByte()), o.asTensor().getByte());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
@@ -460,14 +460,14 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
       }
       index.set(0);
       if (isScalar) {
-        assertTrue(predicate.test(o.data().getByte()));
+        assertTrue(predicate.test(o.asTensor().getByte()));
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.data().getByte())));
+            .forEachIndexed((idx, f) -> assertTrue(predicate.test(o.asTensor().getByte())));
       }
     } else {
-      fail("Unexpected DataType: " + dtype);
+      fail("Unexpected Class: " + type);
     }
   }
 
@@ -482,13 +482,13 @@ public void evaluate(String[] expected, Output<TString> input) {
     AtomicInteger index = new AtomicInteger();
     if (debug) {
       input
-          .data()
+          .asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %s\n", index.getAndIncrement(), f.getObject()));
     }
     index.set(0);
     input
-        .data()
+        .asTensor()
         .scalars()
         .forEach(f -> assertEquals(expected[index.getAndIncrement()], f.getObject()));
   }
@@ -504,13 +504,13 @@ public void evaluate(Boolean[] expected, Output<TBool> input) {
     AtomicInteger index = new AtomicInteger();
     if (debug) {
       input
-          .data()
+          .asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %b\n", index.getAndIncrement(), f.getBoolean()));
     }
     index.set(0);
     input
-        .data()
+        .asTensor()
         .scalars()
         .forEach(f -> assertEquals(expected[index.getAndIncrement()], f.getBoolean()));
   }
@@ -522,184 +522,184 @@ public <T extends TType> void evaluate(Output<T> expected, Output<T> input) {
         : String.format(
             "expected shape (%s) != to input shape (%s)",
             expected.shape().toString(), input.shape().toString());
-    DataType<T> dtype = input.asOutput().dataType();
+    Class<T> type = input.asOutput().type();
     boolean isScalar = input.shape().equals(Shape.scalar());
-    if (dtype == TFloat32.DTYPE) {
+    if (type == TFloat32.class) {
       Output<TFloat32> x = (Output<TFloat32>) expected;
       Output<TFloat32> o = (Output<TFloat32>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
         if (isScalar) {
-          System.out.printf("0). %f <==> %f\n", x.data().getFloat(), o.data().getFloat());
+          System.out.printf("0). %f <==> %f\n", x.asTensor().getFloat(), o.asTensor().getFloat());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
                       System.out.printf(
                           "%d). %f <==> %f\n",
-                          index.getAndIncrement(), x.data().getFloat(idx), f.getFloat()));
+                          index.getAndIncrement(), x.asTensor().getFloat(idx), f.getFloat()));
         }
       }
       index.set(0);
       if (isScalar) {
-        assertEquals(x.data().getFloat(), o.data().getFloat(), epsilon);
+        assertEquals(x.asTensor().getFloat(), o.asTensor().getFloat(), epsilon);
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEachIndexed(
-                (idx, f) -> assertEquals(x.data().getFloat(idx), f.getFloat(), epsilon));
+                (idx, f) -> assertEquals(x.asTensor().getFloat(idx), f.getFloat(), epsilon));
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (type == TFloat64.class) {
       Output<TFloat64> x = (Output<TFloat64>) expected;
       Output<TFloat64> o = (Output<TFloat64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
         if (isScalar) {
-          System.out.printf("0). %f <==> %f\n", x.data().getDouble(), o.data().getDouble());
+          System.out.printf("0). %f <==> %f\n", x.asTensor().getDouble(), o.asTensor().getDouble());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
                       System.out.printf(
                           "%d). %f <==> %f\n",
-                          index.getAndIncrement(), x.data().getDouble(idx), f.getDouble()));
+                          index.getAndIncrement(), x.asTensor().getDouble(idx), f.getDouble()));
         }
       }
       index.set(0);
       if (isScalar) {
-        assertEquals(x.data().getDouble(), o.data().getDouble(), epsilon);
+        assertEquals(x.asTensor().getDouble(), o.asTensor().getDouble(), epsilon);
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
             .forEachIndexed(
-                (idx, f) -> assertEquals(x.data().getDouble(idx), f.getDouble(), epsilon));
+                (idx, f) -> assertEquals(x.asTensor().getDouble(idx), f.getDouble(), epsilon));
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (type == TInt32.class) {
       Output<TInt32> x = (Output<TInt32>) expected;
       Output<TInt32> o = (Output<TInt32>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
         if (isScalar) {
-          System.out.printf("0). %d <==> %d\n", x.data().getInt(), o.data().getInt());
+          System.out.printf("0). %d <==> %d\n", x.asTensor().getInt(), o.asTensor().getInt());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
                       System.out.printf(
                           "%d). %d <==> %d\n",
-                          index.getAndIncrement(), x.data().getInt(idx), f.getInt()));
+                          index.getAndIncrement(), x.asTensor().getInt(idx), f.getInt()));
         }
       }
       index.set(0);
       if (isScalar) {
-        assertEquals(x.data().getInt(), o.data().getInt());
+        assertEquals(x.asTensor().getInt(), o.asTensor().getInt());
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertEquals(x.data().getInt(idx), f.getInt()));
+            .forEachIndexed((idx, f) -> assertEquals(x.asTensor().getInt(idx), f.getInt()));
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (type == TInt64.class) {
       Output<TInt64> x = (Output<TInt64>) expected;
       Output<TInt64> o = (Output<TInt64>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
         if (isScalar) {
-          System.out.printf("0). %d  <==> %d\n", x.data().getLong(), o.data().getLong());
+          System.out.printf("0). %d  <==> %d\n", x.asTensor().getLong(), o.asTensor().getLong());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
                       System.out.printf(
                           "%d). %d  <==> %d\n",
-                          index.getAndIncrement(), x.data().getLong(idx), f.getLong()));
+                          index.getAndIncrement(), x.asTensor().getLong(idx), f.getLong()));
         }
       }
       index.set(0);
       if (isScalar) {
-        assertEquals(x.data().getLong(), o.data().getLong());
+        assertEquals(x.asTensor().getLong(), o.asTensor().getLong());
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertEquals(x.data().getLong(idx), f.getLong()));
+            .forEachIndexed((idx, f) -> assertEquals(x.asTensor().getLong(idx), f.getLong()));
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (type == TUint8.class) {
       Output<TUint8> x = (Output<TUint8>) expected;
       Output<TUint8> o = (Output<TUint8>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
         if (isScalar) {
-          System.out.printf("0). %x  <==> %x\n", x.data().getByte(), o.data().getByte());
+          System.out.printf("0). %x  <==> %x\n", x.asTensor().getByte(), o.asTensor().getByte());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
                       System.out.printf(
                           "%d). %x  <==> %x\n",
-                          index.getAndIncrement(), x.data().getByte(idx), f.getByte()));
+                          index.getAndIncrement(), x.asTensor().getByte(idx), f.getByte()));
         }
       }
       index.set(0);
       if (isScalar) {
-        assertEquals(x.data().getByte(), o.data().getByte());
+        assertEquals(x.asTensor().getByte(), o.asTensor().getByte());
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertEquals(x.data().getByte(idx), f.getByte()));
+            .forEachIndexed((idx, f) -> assertEquals(x.asTensor().getByte(idx), f.getByte()));
       }
-    } else if (dtype == TString.DTYPE) {
+    } else if (type == TString.class) {
       Output<TString> x = (Output<TString>) expected;
       Output<TString> o = (Output<TString>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
         if (isScalar) {
-          System.out.printf("0). %s  <==> %s\n", x.data().getObject(), o.data().getObject());
+          System.out.printf("0). %s  <==> %s\n", x.asTensor().getObject(), o.asTensor().getObject());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
                       System.out.printf(
                           "%d). %s  <==> %s\n",
-                          index.getAndIncrement(), x.data().getObject(idx), f.getObject()));
+                          index.getAndIncrement(), x.asTensor().getObject(idx), f.getObject()));
         }
       }
       index.set(0);
       if (isScalar) {
-        assertEquals(x.data().getObject(), o.data().getObject());
+        assertEquals(x.asTensor().getObject(), o.asTensor().getObject());
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertEquals(x.data().getObject(idx), f.getObject()));
+            .forEachIndexed((idx, f) -> assertEquals(x.asTensor().getObject(idx), f.getObject()));
       }
-    } else if (dtype == TBool.DTYPE) {
+    } else if (type == TBool.class) {
       Output<TBool> x = (Output<TBool>) expected;
       Output<TBool> o = (Output<TBool>) input;
       AtomicInteger index = new AtomicInteger();
       if (debug) {
         if (isScalar) {
-          System.out.printf("0). %b  <==> %b\n", x.data().getBoolean(), o.data().getBoolean());
+          System.out.printf("0). %b  <==> %b\n", x.asTensor().getBoolean(), o.asTensor().getBoolean());
         } else {
-          o.data()
+          o.asTensor()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
                       System.out.printf(
                           "%d). %b  <==> %b\n",
-                          index.getAndIncrement(), x.data().getBoolean(idx), f.getBoolean()));
+                          index.getAndIncrement(), x.asTensor().getBoolean(idx), f.getBoolean()));
         }
       }
       index.set(0);
       if (isScalar) {
-        assertEquals(x.data().getBoolean(), o.data().getBoolean());
+        assertEquals(x.asTensor().getBoolean(), o.asTensor().getBoolean());
       } else {
-        o.data()
+        o.asTensor()
             .scalars()
-            .forEachIndexed((idx, f) -> assertEquals(x.data().getBoolean(idx), f.getBoolean()));
+            .forEachIndexed((idx, f) -> assertEquals(x.asTensor().getBoolean(idx), f.getBoolean()));
       }
     }
   }
@@ -707,51 +707,51 @@ public <T extends TType> void evaluate(Output<T> expected, Output<T> input) {
   /** {@inheritDoc} */
   @Override
   public <T extends TType> void print(PrintWriter writer, Output<T> input) {
-    DataType<T> dtype = input.asOutput().dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> type = input.asOutput().type();
+    if (type == TFloat32.class) {
       Output<TFloat32> o = (Output<TFloat32>) input;
       AtomicInteger index = new AtomicInteger();
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (type == TFloat64.class) {
       Output<TFloat64> o = (Output<TFloat64>) input;
       AtomicInteger index = new AtomicInteger();
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (type == TInt32.class) {
       Output<TInt32> o = (Output<TInt32>) input;
       AtomicInteger index = new AtomicInteger();
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (type == TInt64.class) {
       Output<TInt64> o = (Output<TInt64>) input;
       AtomicInteger index = new AtomicInteger();
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (type == TUint8.class) {
       Output<TUint8> o = (Output<TUint8>) input;
       AtomicInteger index = new AtomicInteger();
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %x\n", index.getAndIncrement(), f.getByte()));
-    } else if (dtype == TString.DTYPE) {
+    } else if (type == TString.class) {
       Output<TString> o = (Output<TString>) input;
       AtomicInteger index = new AtomicInteger();
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %s\n", index.getAndIncrement(), f.getObject()));
-    } else if (dtype == TBool.DTYPE) {
+    } else if (type == TBool.class) {
       Output<TBool> o = (Output<TBool>) input;
       AtomicInteger index = new AtomicInteger();
-      o.data()
+      o.asTensor()
           .scalars()
           .forEach(f -> System.out.printf("%d). %b\n", index.getAndIncrement(), f.getBoolean()));
     } else {
-      writer.println("Unexpected DataType: " + dtype);
+      writer.println("Unexpected Class: " + type);
     }
     writer.flush();
   }
diff --git a/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/GraphTestSession.java b/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/GraphTestSession.java
index 33ddec6dce3..d2e856e9f22 100644
--- a/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/GraphTestSession.java
+++ b/tensorflow-framework/src/test/java/org/tensorflow/framework/utils/GraphTestSession.java
@@ -104,89 +104,84 @@ public void run(Op op) {
   /** {@inheritDoc} */
   @Override
   public <T extends TNumber> void evaluate(double expected, Operand<T> input) {
-    DataType<T> dtype = input.asOutput().dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> dtype = input.asOutput().type();
+    if (dtype == TFloat32.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TFloat32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+        try (TFloat32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
         }
       }
       index.set(0);
-      try (Tensor<TFloat32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
-        result.data().scalars().forEach(f -> assertEquals((float) expected, f.getFloat(), epsilon));
+      try (TFloat32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
+        result.scalars().forEach(f -> assertEquals((float) expected, f.getFloat(), epsilon));
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (dtype == TFloat64.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TFloat64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+        try (TFloat64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
         }
       }
       index.set(0);
-      try (Tensor<TFloat64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
-        result.data().scalars().forEach(f -> assertEquals(expected, f.getDouble(), epsilon));
+      try (TFloat64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
+        result.scalars().forEach(f -> assertEquals(expected, f.getDouble(), epsilon));
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (dtype == TInt32.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TInt32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+        try (TInt32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
         }
       }
       index.set(0);
-      try (Tensor<TInt32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
-        result.data().scalars().forEach(f -> assertEquals((int) expected, f.getInt()));
+      try (TInt32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
+        result.scalars().forEach(f -> assertEquals((int) expected, f.getInt()));
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (dtype == TInt64.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TInt64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+        try (TInt64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
         }
       }
       index.set(0);
-      try (Tensor<TInt64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
-        result.data().scalars().forEach(f -> assertEquals((long) expected, f.getLong()));
+      try (TInt64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
+        result.scalars().forEach(f -> assertEquals((long) expected, f.getLong()));
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (dtype == TUint8.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TUint8> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+        try (TUint8 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getByte()));
         }
       }
       index.set(0);
-      try (Tensor<TUint8> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
-        result.data().scalars().forEach(f -> assertEquals((long) expected, f.getByte()));
+      try (TUint8 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
+        result.scalars().forEach(f -> assertEquals((long) expected, f.getByte()));
       }
     } else {
-      fail("Unexpected DataType: " + dtype);
+      fail("Unexpected Class: " + dtype);
     }
   }
 
@@ -198,223 +193,203 @@ public <T extends TNumber> void evaluate(Number[] expected, Output<T> input) {
         expected.length,
         size,
         () -> String.format("expected length (%d) != to input length (%d)", expected.length, size));
-    DataType<T> dtype = input.asOutput().dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> dtype = input.asOutput().type();
+    if (dtype == TFloat32.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TFloat32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+        try (TFloat32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
         }
       }
       index.set(0);
-      try (Tensor<TFloat32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+      try (TFloat32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(
                 f ->
                     assertEquals(
                         expected[index.getAndIncrement()].floatValue(), f.getFloat(), epsilon));
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (dtype == TFloat64.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TFloat64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+        try (TFloat64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
         }
       }
       index.set(0);
-      try (Tensor<TFloat64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+      try (TFloat64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(
                 f ->
                     assertEquals(
                         expected[index.getAndIncrement()].doubleValue(), f.getDouble(), epsilon));
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (dtype == TInt32.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TInt32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+        try (TInt32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
         }
       }
       index.set(0);
-      try (Tensor<TInt32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(f -> assertEquals(expected[index.getAndIncrement()].intValue(), f.getInt()));
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (dtype == TInt64.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TInt64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+        try (TInt64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
         }
       }
       index.set(0);
-      try (Tensor<TInt64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(f -> assertEquals(expected[index.getAndIncrement()].longValue(), f.getLong()));
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (dtype == TUint8.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TUint8> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+        try (TUint8 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getByte()));
         }
       }
       index.set(0);
-      try (Tensor<TUint8> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+      try (TUint8 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(f -> assertEquals(expected[index.getAndIncrement()].longValue(), f.getByte()));
       }
     } else {
-      fail("Unexpected DataType: " + dtype);
+      fail("Unexpected Class: " + dtype);
     }
   }
 
   /** {@inheritDoc} */
   @Override
   public <T extends TType> void evaluate(FloatNdArray expected, Output<T> input) {
-    DataType<T> dtype = input.asOutput().dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> dtype = input.asOutput().type();
+    if (dtype == TFloat32.class) {
       AtomicLong index = new AtomicLong();
       if (debug) {
-        try (Tensor<TFloat32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+        try (TFloat32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
         }
       }
       index.set(0);
-      try (Tensor<TFloat32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+      try (TFloat32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(
                 f ->
                     assertEquals(
                         expected.getFloat(index.getAndIncrement()), f.getFloat(), epsilon));
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (dtype == TFloat64.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TFloat64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+        try (TFloat64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
         }
       }
       index.set(0);
-      try (Tensor<TFloat64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+      try (TFloat64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(
                 f ->
                     assertEquals(
                         expected.getFloat(index.getAndIncrement()), f.getDouble(), epsilon));
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (dtype == TInt32.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TInt32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+        try (TInt32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
         }
       }
       index.set(0);
-      try (Tensor<TInt32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(
                 f -> assertEquals((int) expected.getFloat(index.getAndIncrement()), f.getInt()));
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (dtype == TInt64.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TInt64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+        try (TInt64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
         }
       }
       index.set(0);
-      try (Tensor<TInt64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(
                 f -> assertEquals((long) expected.getFloat(index.getAndIncrement()), f.getLong()));
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (dtype == TUint8.class) {
       AtomicInteger index = new AtomicInteger();
       if (debug) {
-        try (Tensor<TUint8> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+        try (TUint8 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           result
-              .data()
               .scalars()
               .forEach(f -> System.out.printf("%d). %d\n", index.getAndIncrement(), f.getByte()));
         }
       }
       index.set(0);
-      try (Tensor<TUint8> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+      try (TUint8 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(
                 f -> assertEquals((long) expected.getFloat(index.getAndIncrement()), f.getByte()));
       }
     } else {
-      fail("Unexpected DataType: " + dtype);
+      fail("Unexpected Class: " + dtype);
     }
   }
 
@@ -428,19 +403,17 @@ public void evaluate(String[] expected, Output<TString> input) {
         () -> String.format("expected length (%d) != to input length (%d)", expected.length, size));
     AtomicInteger index = new AtomicInteger();
     if (debug) {
-      try (Tensor<TString> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE)) {
+      try (TString result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(f -> System.out.printf("%d). %s\n", index.getAndIncrement(), f.getObject()));
       }
     }
     index.set(0);
-    try (Tensor<TString> result =
-        this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE)) {
+    try (TString result =
+        this.getGraphSession().runner().fetch(input).run().get(0)) {
       result
-          .data()
           .scalars()
           .forEach(f -> assertEquals(expected[index.getAndIncrement()], f.getObject()));
     }
@@ -456,19 +429,17 @@ public void evaluate(Boolean[] expected, Output<TBool> input) {
         () -> String.format("expected length (%d) != to input length (%d)", expected.length, size));
     AtomicInteger index = new AtomicInteger();
     if (debug) {
-      try (Tensor<TBool> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TBool.DTYPE)) {
+      try (TBool result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         result
-            .data()
             .scalars()
             .forEach(f -> System.out.printf("%d). %b\n", index.getAndIncrement(), f.getObject()));
       }
     }
     index.set(0);
-    try (Tensor<TBool> result =
-        this.getGraphSession().runner().fetch(input).run().get(0).expect(TBool.DTYPE)) {
+    try (TBool result =
+        this.getGraphSession().runner().fetch(input).run().get(0)) {
       result
-          .data()
           .scalars()
           .forEach(f -> assertEquals(expected[index.getAndIncrement()], f.getObject()));
     }
@@ -482,331 +453,313 @@ public <T extends TType> void evaluate(Output<T> expected, Output<T> input) {
             "expected shape (%s) != to input shape (%s)",
             expected.shape().toString(), input.shape().toString());
     AtomicInteger index = new AtomicInteger();
-    DataType<T> dtype = input.asOutput().dataType();
-    if (!dtype.equals(expected.dataType())) {
+    Class<T> dtype = input.asOutput().type();
+    if (!dtype.equals(expected.type())) {
       throw new IllegalArgumentException(
           String.format(
               "Both data type must be equal, inout = %s, expected = %s",
-              dtype, expected.dataType()));
+              dtype, expected.type()));
     }
     boolean isScalar = input.shape().equals(Shape.scalar());
-    if (dtype == TFloat32.DTYPE) {
+    if (dtype == TFloat32.class) {
       final Output<TFloat32> finalExpected = (Output<TFloat32>) expected;
       if (debug) {
-        try (Tensor<TFloat32> result =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE);
-            Tensor<TFloat32> expectedResult =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+        try (TFloat32 result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TFloat32 expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %f <==> %f\n", expectedResult.data().getFloat(), result.data().getFloat());
+                "0). %f <==> %f\n", expectedResult.getFloat(), result.getFloat());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
                         System.out.printf(
                             "%d). %f <==> %f\n",
                             index.getAndIncrement(),
-                            finalExpected.data().getFloat(idx),
+                            finalExpected.asTensor().getFloat(idx),
                             f.getFloat()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TFloat32> result =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE);
-          Tensor<TFloat32> expectedResult =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+      try (TFloat32 result =
+              this.getGraphSession().runner().fetch(input).run().get(0);
+          TFloat32 expectedResult =
+              this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getFloat(), result.data().getFloat(), epsilon);
+          assertEquals(expectedResult.getFloat(), result.getFloat(), epsilon);
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
-                      assertEquals(expectedResult.data().getFloat(idx), f.getFloat(), epsilon));
+                      assertEquals(expectedResult.getFloat(idx), f.getFloat(), epsilon));
         }
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (dtype == TFloat64.class) {
       final Output<TFloat64> finalExpected = (Output<TFloat64>) expected;
       if (debug) {
-        try (Tensor<TFloat64> result =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE);
-            Tensor<TFloat64> expectedResult =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+        try (TFloat64 result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TFloat64 expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %f <==> %f\n", expectedResult.data().getDouble(), result.data().getDouble());
+                "0). %f <==> %f\n", expectedResult.getDouble(), result.getDouble());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
                         System.out.printf(
                             "%d). %f <==> %f\n",
                             index.getAndIncrement(),
-                            finalExpected.data().getDouble(idx),
+                            finalExpected.asTensor().getDouble(idx),
                             f.getDouble()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TFloat64> result =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE);
-          Tensor<TFloat64> expectedResult =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+      try (TFloat64 result =
+              this.getGraphSession().runner().fetch(input).run().get(0);
+          TFloat64 expectedResult =
+              this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getDouble(), result.data().getDouble(), epsilon);
+          assertEquals(expectedResult.getDouble(), result.getDouble(), epsilon);
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
-                      assertEquals(expectedResult.data().getDouble(idx), f.getDouble(), epsilon));
+                      assertEquals(expectedResult.getDouble(idx), f.getDouble(), epsilon));
         }
       }
-    } else if (dtype == TFloat16.DTYPE) {
+    } else if (dtype == TFloat16.class) {
       final Output<TFloat16> finalExpected = (Output<TFloat16>) expected;
       if (debug) {
-        try (Tensor<TFloat16> result =
-                     this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat16.DTYPE);
-             Tensor<TFloat16> expectedResult =
-                     this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat16.DTYPE)) {
+        try (TFloat16 result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TFloat16 expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                    "0). %f <==> %f\n", expectedResult.data().getFloat(), result.data().getFloat());
+                "0). %f <==> %f\n", expectedResult.getFloat(), result.getFloat());
           } else {
             result
-                    .data()
-                    .scalars()
-                    .forEachIndexed(
-                            (idx, f) ->
-                                    System.out.printf(
-                                            "%d). %f <==> %f\n",
-                                            index.getAndIncrement(),
-                                            finalExpected.data().getFloat(idx),
-                                            f.getFloat()));
+                .scalars()
+                .forEachIndexed(
+                    (idx, f) ->
+                        System.out.printf(
+                            "%d). %f <==> %f\n",
+                            index.getAndIncrement(),
+                            finalExpected.asTensor().getFloat(idx),
+                            f.getFloat()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TFloat16> result =
-                   this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat16.DTYPE);
-           Tensor<TFloat16> expectedResult =
-                   this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat16.DTYPE)) {
+      try (TFloat16 result =
+              this.getGraphSession().runner().fetch(input).run().get(0);
+          TFloat16 expectedResult =
+              this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getFloat(), result.data().getFloat(), epsilon);
+          assertEquals(expectedResult.getFloat(), result.getFloat(), epsilon);
         } else {
           result
-                  .data()
-                  .scalars()
-                  .forEachIndexed(
-                          (idx, f) ->
-                                  assertEquals(expectedResult.data().getFloat(idx), f.getFloat(), epsilon));
+              .scalars()
+              .forEachIndexed(
+                  (idx, f) ->
+                      assertEquals(expectedResult.getFloat(idx), f.getFloat(), epsilon));
         }
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (dtype == TInt32.class) {
       final Output<TInt32> finalExpected = (Output<TInt32>) expected;
       if (debug) {
-        try (Tensor<TInt32> result =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE);
-            Tensor<TInt32> expectedResult =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+        try (TInt32 result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TInt32 expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %d <==> %d\n", expectedResult.data().getInt(), result.data().getInt());
+                "0). %d <==> %d\n", expectedResult.getInt(), result.getInt());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
                         System.out.printf(
                             "%d). %d <==> %d\n",
-                            index.getAndIncrement(), finalExpected.data().getInt(idx), f.getInt()));
+                            index.getAndIncrement(), finalExpected.asTensor().getInt(idx), f.getInt()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TInt32> result =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE);
-          Tensor<TInt32> expectedResult =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+              this.getGraphSession().runner().fetch(input).run().get(0);
+          TInt32 expectedResult =
+              this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getInt(), result.data().getInt(), epsilon);
+          assertEquals(expectedResult.getInt(), result.getInt(), epsilon);
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
-                  (idx, f) -> assertEquals(expectedResult.data().getInt(idx), f.getInt(), epsilon));
+                  (idx, f) -> assertEquals(expectedResult.getInt(idx), f.getInt(), epsilon));
         }
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (dtype == TInt64.class) {
       final Output<TInt64> finalExpected = (Output<TInt64>) expected;
       if (debug) {
-        try (Tensor<TInt64> result =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE);
-            Tensor<TInt64> expectedResult =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+        try (TInt64 result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TInt64 expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %d <==> %d\n", expectedResult.data().getLong(), result.data().getLong());
+                "0). %d <==> %d\n", expectedResult.getLong(), result.getLong());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
                         System.out.printf(
                             "%d). %d <==> %d\n",
                             index.getAndIncrement(),
-                            finalExpected.data().getLong(idx),
+                            finalExpected.asTensor().getLong(idx),
                             f.getLong()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TInt64> result =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE);
-          Tensor<TInt64> expectedResult =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+              this.getGraphSession().runner().fetch(input).run().get(0);
+          TInt64 expectedResult =
+              this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getLong(), result.data().getLong(), epsilon);
+          assertEquals(expectedResult.getLong(), result.getLong(), epsilon);
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
-                      assertEquals(expectedResult.data().getLong(idx), f.getLong(), epsilon));
+                      assertEquals(expectedResult.getLong(idx), f.getLong(), epsilon));
         }
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (dtype == TUint8.class) {
       final Output<TUint8> finalExpected = (Output<TUint8>) expected;
       if (debug) {
-        try (Tensor<TUint8> result =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE);
-            Tensor<TUint8> expectedResult =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+        try (TUint8 result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TUint8 expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %d <==> %d\n", expectedResult.data().getByte(), result.data().getByte());
+                "0). %d <==> %d\n", expectedResult.getByte(), result.getByte());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
                         System.out.printf(
                             "%d). %d <==> %d\n",
                             index.getAndIncrement(),
-                            finalExpected.data().getByte(idx),
+                            finalExpected.asTensor().getByte(idx),
                             f.getByte()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TUint8> result =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE);
-          Tensor<TUint8> expectedResult =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+      try (TUint8 result =
+              this.getGraphSession().runner().fetch(input).run().get(0);
+          TUint8 expectedResult =
+              this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getByte(), result.data().getByte(), epsilon);
+          assertEquals(expectedResult.getByte(), result.getByte(), epsilon);
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
-                      assertEquals(expectedResult.data().getByte(idx), f.getByte(), epsilon));
+                      assertEquals(expectedResult.getByte(idx), f.getByte(), epsilon));
         }
       }
-    } else if (dtype == TBool.DTYPE) {
+    } else if (dtype == TBool.class) {
       final Output<TBool> finalExpected = (Output<TBool>) expected;
       if (debug) {
-        try (Tensor<TBool> result =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TBool.DTYPE);
-            Tensor<TBool> expectedResult =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TBool.DTYPE)) {
+        try (TBool result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TBool expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %b <==> %b\n", expectedResult.data().getBoolean(), result.data().getBoolean());
+                "0). %b <==> %b\n", expectedResult.getBoolean(), result.getBoolean());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
                         System.out.printf(
                             "%d). %b <==> %b\n",
                             index.getAndIncrement(),
-                            finalExpected.data().getBoolean(idx),
+                            finalExpected.asTensor().getBoolean(idx),
                             f.getBoolean()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TBool> result =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TBool.DTYPE);
-          Tensor<TBool> expectedResult =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TBool.DTYPE)) {
+      try (TBool result = this.getGraphSession().runner().fetch(input).run().get(0);
+          TBool expectedResult = this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getBoolean(), result.data().getBoolean());
+          assertEquals(expectedResult.getBoolean(), result.getBoolean());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
-                  (idx, f) -> assertEquals(expectedResult.data().getBoolean(idx), f.getBoolean()));
+                  (idx, f) -> assertEquals(expectedResult.getBoolean(idx), f.getBoolean()));
         }
       }
-    } else if (dtype == TString.DTYPE) {
+    } else if (dtype == TString.class) {
       final Output<TString> finalExpected = (Output<TString>) expected;
       if (debug) {
-        try (Tensor<TString> result =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE);
-            Tensor<TString> expectedResult =
-                this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE)) {
+        try (TString result =
+                this.getGraphSession().runner().fetch(input).run().get(0);
+            TString expectedResult =
+                this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %s <==> %s\n", expectedResult.data().getObject(), result.data().getObject());
+                "0). %s <==> %s\n", expectedResult.getObject(), result.getObject());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
                         System.out.printf(
                             "%d). %s <==> %s\n",
                             index.getAndIncrement(),
-                            finalExpected.data().getObject(idx),
+                            finalExpected.asTensor().getObject(idx),
                             f.getObject()));
           }
         }
       }
       index.set(0);
-      try (Tensor<TString> result =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE);
-          Tensor<TString> expectedResult =
-              this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE)) {
+      try (TString result =
+              this.getGraphSession().runner().fetch(input).run().get(0);
+          TString expectedResult =
+              this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertEquals(expectedResult.data().getObject(), result.data().getObject());
+          assertEquals(expectedResult.getObject(), result.getObject());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
-                  (idx, f) -> assertEquals(expectedResult.data().getObject(idx), f.getObject()));
+                  (idx, f) -> assertEquals(expectedResult.getObject(idx), f.getObject()));
         }
       }
     } else {
-      fail("Unexpected DataType: " + dtype);
+      fail("Unexpected Class: " + dtype);
     }
   }
 
@@ -816,15 +769,14 @@ public void evaluateString(Output<TString> input, Predicate<String> predicate) {
     boolean isScalar = input.shape().equals(Shape.scalar());
     AtomicInteger index = new AtomicInteger();
     if (debug) {
-      try (Tensor<TString> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE)) {
+      try (TString result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
           System.out.printf(
               "0). %b <==> %s\n",
-              predicate.test(result.data().getObject()), result.data().getObject());
+              predicate.test(result.getObject()), result.getObject());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) ->
@@ -835,13 +787,12 @@ public void evaluateString(Output<TString> input, Predicate<String> predicate) {
       }
     }
     index.set(0);
-    try (Tensor<TString> result =
-        this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE)) {
+    try (TString result =
+        this.getGraphSession().runner().fetch(input).run().get(0)) {
       if (isScalar) {
-        assertTrue(predicate.test(result.data().getObject()));
+        assertTrue(predicate.test(result.getObject()));
       } else {
         result
-            .data()
             .scalars()
             .forEachIndexed((idx, s) -> assertTrue(predicate.test(s.getObject())));
       }
@@ -852,19 +803,18 @@ public void evaluateString(Output<TString> input, Predicate<String> predicate) {
   @Override
   public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predicate) {
     AtomicInteger index = new AtomicInteger();
-    DataType<T> dtype = input.asOutput().dataType();
+    Class<T> dtype = input.asOutput().type();
     boolean isScalar = input.shape().equals(Shape.scalar());
-    if (dtype == TFloat32.DTYPE) {
+    if (dtype == TFloat32.class) {
       if (debug) {
-        try (Tensor<TFloat32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+        try (TFloat32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
                 "0). %b <==> %f\n",
-                predicate.test(result.data().getFloat()), result.data().getFloat());
+                predicate.test(result.getFloat()), result.getFloat());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
@@ -875,28 +825,26 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
         }
       }
       index.set(0);
-      try (Tensor<TFloat32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+      try (TFloat32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertTrue(predicate.test(result.data().getFloat()));
+          assertTrue(predicate.test(result.getFloat()));
         } else {
           result
-              .data()
               .scalars()
-              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.data().getFloat())));
+              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.getFloat())));
         }
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (dtype == TFloat64.class) {
       if (debug) {
-        try (Tensor<TFloat64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+        try (TFloat64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
                 "0). %b <==> %f\n",
-                predicate.test(result.data().getDouble()), result.data().getDouble());
+                predicate.test(result.getDouble()), result.getDouble());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
@@ -907,27 +855,25 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
         }
       }
       index.set(0);
-      try (Tensor<TFloat64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+      try (TFloat64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertTrue(predicate.test(result.data().getDouble()));
+          assertTrue(predicate.test(result.getDouble()));
         } else {
           result
-              .data()
               .scalars()
-              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.data().getDouble())));
+              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.getDouble())));
         }
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (dtype == TInt32.class) {
       if (debug) {
-        try (Tensor<TInt32> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+        try (TInt32 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
-                "0). %b <==> %d\n", predicate.test(result.data().getInt()), result.data().getInt());
+                "0). %b <==> %d\n", predicate.test(result.getInt()), result.getInt());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
@@ -938,28 +884,26 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
         }
       }
       index.set(0);
-      try (Tensor<TInt32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertTrue(predicate.test(result.data().getInt()));
+          assertTrue(predicate.test(result.getInt()));
         } else {
           result
-              .data()
               .scalars()
-              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.data().getInt())));
+              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.getInt())));
         }
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (dtype == TInt64.class) {
       if (debug) {
-        try (Tensor<TInt64> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+        try (TInt64 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
                 "0). %b <==> %d\n",
-                predicate.test(result.data().getLong()), result.data().getLong());
+                predicate.test(result.getLong()), result.getLong());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
@@ -970,28 +914,26 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
         }
       }
       index.set(0);
-      try (Tensor<TInt64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertTrue(predicate.test(result.data().getLong()));
+          assertTrue(predicate.test(result.getLong()));
         } else {
           result
-              .data()
               .scalars()
-              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.data().getLong())));
+              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.getLong())));
         }
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (dtype == TUint8.class) {
       if (debug) {
-        try (Tensor<TUint8> result =
-            this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+        try (TUint8 result =
+            this.getGraphSession().runner().fetch(input).run().get(0)) {
           if (isScalar) {
             System.out.printf(
                 "0). %b <==> %d\n",
-                predicate.test(result.data().getByte()), result.data().getByte());
+                predicate.test(result.getByte()), result.getByte());
           } else {
             result
-                .data()
                 .scalars()
                 .forEachIndexed(
                     (idx, f) ->
@@ -1002,19 +944,18 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
         }
       }
       index.set(0);
-      try (Tensor<TUint8> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+      try (TUint8 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          assertTrue(predicate.test(result.data().getByte()));
+          assertTrue(predicate.test(result.getByte()));
         } else {
           result
-              .data()
               .scalars()
-              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.data().getByte())));
+              .forEachIndexed((idx, f) -> assertTrue(predicate.test(result.getByte())));
         }
       }
     } else {
-      fail("Unexpected DataType: " + dtype);
+      fail("Unexpected Class: " + dtype);
     }
   }
 
@@ -1023,119 +964,112 @@ public <T extends TType> void evaluate(Output<T> input, Predicate<Number> predic
   public <T extends TType> void print(PrintWriter writer, Output<T> input) {
     boolean isScalar = input.asOutput().shape().size() == 1;
 
-    DataType<T> dtype = input.dataType();
-    if (dtype == TFloat32.DTYPE) {
+    Class<T> dtype = input.type();
+    if (dtype == TFloat32.class) {
       AtomicInteger index = new AtomicInteger();
-      try (Tensor<TFloat32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat32.DTYPE)) {
+      try (TFloat32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
-          writer.printf("%d). %f\n", index.getAndIncrement(), result.data().getFloat());
+          writer.printf("%d). %f\n", index.getAndIncrement(), result.getFloat());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) -> writer.printf("%d). %f\n", index.getAndIncrement(), f.getFloat()));
         }
       }
-    } else if (dtype == TFloat64.DTYPE) {
+    } else if (dtype == TFloat64.class) {
       AtomicInteger index = new AtomicInteger();
 
-      try (Tensor<TFloat64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TFloat64.DTYPE)) {
+      try (TFloat64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
           writer.printf(
-              "%d). %f\n", index.getAndIncrement(), ((Output<TFloat64>) input).data().getDouble());
+              "%d). %f\n", index.getAndIncrement(), ((Output<TFloat64>) input).asTensor().getDouble());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) -> writer.printf("%d). %f\n", index.getAndIncrement(), f.getDouble()));
         }
       }
-    } else if (dtype == TInt32.DTYPE) {
+    } else if (dtype == TInt32.class) {
       AtomicInteger index = new AtomicInteger();
 
-      try (Tensor<TInt32> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt32.DTYPE)) {
+      try (TInt32 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
           writer.printf(
-              "%d). %d\n", index.getAndIncrement(), ((Output<TInt32>) input).data().getInt());
+              "%d). %d\n", index.getAndIncrement(), ((Output<TInt32>) input).asTensor().getInt());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) -> writer.printf("%d). %d\n", index.getAndIncrement(), f.getInt()));
         }
       }
-    } else if (dtype == TInt64.DTYPE) {
+    } else if (dtype == TInt64.class) {
       AtomicInteger index = new AtomicInteger();
 
-      try (Tensor<TInt64> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TInt64.DTYPE)) {
+      try (TInt64 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
           writer.printf(
-              "%d). %d\n", index.getAndIncrement(), ((Output<TInt64>) input).data().getLong());
+              "%d). %d\n", index.getAndIncrement(), ((Output<TInt64>) input).asTensor().getLong());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) -> writer.printf("%d). %d\n", index.getAndIncrement(), f.getLong()));
         }
       }
-    } else if (dtype == TUint8.DTYPE) {
+    } else if (dtype == TUint8.class) {
       AtomicInteger index = new AtomicInteger();
 
-      try (Tensor<TUint8> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TUint8.DTYPE)) {
+      try (TUint8 result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
           writer.printf(
-              "%d). %x\n", index.getAndIncrement(), ((Output<TUint8>) input).data().getByte());
+              "%d). %x\n", index.getAndIncrement(), ((Output<TUint8>) input).asTensor().getByte());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) -> writer.printf("%d). %x\n", index.getAndIncrement(), f.getByte()));
         }
       }
-    } else if (dtype == TBool.DTYPE) {
+    } else if (dtype == TBool.class) {
       AtomicInteger index = new AtomicInteger();
 
-      try (Tensor<TBool> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TBool.DTYPE)) {
+      try (TBool result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
           writer.printf(
-              "%d). %b\n", index.getAndIncrement(), ((Output<TBool>) input).data().getBoolean());
+              "%d). %b\n", index.getAndIncrement(), ((Output<TBool>) input).asTensor().getBoolean());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) -> writer.printf("%d). %b\n", index.getAndIncrement(), f.getBoolean()));
         }
       }
-    } else if (dtype == TString.DTYPE) {
+    } else if (dtype == TString.class) {
       AtomicInteger index = new AtomicInteger();
 
-      try (Tensor<TString> result =
-          this.getGraphSession().runner().fetch(input).run().get(0).expect(TString.DTYPE)) {
+      try (TString result =
+          this.getGraphSession().runner().fetch(input).run().get(0)) {
         if (isScalar) {
           writer.printf(
-              "%d). %s\n", index.getAndIncrement(), ((Output<TString>) input).data().getObject());
+              "%d). %s\n", index.getAndIncrement(), ((Output<TString>) input).asTensor().getObject());
         } else {
           result
-              .data()
               .scalars()
               .forEachIndexed(
                   (idx, f) -> writer.printf("%d). %s\n", index.getAndIncrement(), f.getObject()));
         }
       }
     } else {
-      writer.println("Unexpected DataType: " + dtype);
+      writer.println("Unexpected Class: " + dtype);
     }
     writer.flush();
   }