add bin tests

Author: papajohn

datascience/tables.py

@@ -1094,7 +1094,7 @@ def with_relabeling(self, *args):
         warnings.warn("with_relabeling is deprecated; use relabeled", FutureWarning)
         return self.relabeled(*args)
 
-    def bin(self, **vargs):
+    def bin(self, select=None, **vargs):
         """Group values by bin and compute counts per bin by column.
 
         By default, bins are chosen to contain all values in all columns. The
@@ -1105,6 +1105,9 @@ def bin(self, **vargs):
         n+1 columns, where column 0 contains the lower bound of each bin.
 
         Args:
+            ``select`` (columns): Columns to be binned. If None, all columns
+                are binned.
+
             ``bins`` (int or sequence of scalars): If bins is an int,
                 it defines the number of equal-width bins in the given range
                 (10, by default). If bins is a sequence, it defines the bin
@@ -1122,6 +1125,8 @@ def bin(self, **vargs):
                 histogram values will not be equal to 1 unless bins of unity
                 width are chosen; it is not a probability mass function.
         """
+        if select is not None:
+            self = self.select(select)
         if 'normed' in vargs:
             vargs.setdefault('density', vargs.pop('normed'))
         density = vargs.get('density', False)

tests/test_tables.py

@@ -674,6 +674,26 @@ def test_with_relabeling(table):
     z      | 1     | 10
     """)
 
+def test_bin(table):
+    binned = table.bin('count')
+    assert_equal(binned.take(np.arange(5)), """
+    bin  | count count
+    1    | 1
+    1.8  | 0
+    2.6  | 2
+    3.4  | 0
+    4.2  | 0
+    """)
+    binned = table.select([1, 2]).bin(bins=4)
+    assert_equal(binned, """
+    bin  | count count | points count
+    1    | 3           | 3
+    3.25 | 0           | 0
+    5.5  | 0           | 0
+    7.75 | 1           | 1
+    10   | 0           | 0
+    """)
+
 
 ##########
 # Create #