Windows support

[SoftologyPro]

Firstly, is the 8 GPU only for training or for inference too? What GPU/VRAM is needed for running inference in the gradio UI?

Secondly, I am working on getting an install working for windows but the last line of the requirements.txt for udf_ext is not cooperating. Any chance you have a WHL for that? Or an original URL/git repo?

This is the error I get when I manually try and pip install that third_party\voxelize directory

D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize>pip install .
Processing d:\tests\direct3d-s2\direct3d-s2\third_party\voxelize
  Preparing metadata (setup.py) ... done
Building wheels for collected packages: udf-ext
with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'udf-ext'. Discussion can be found at https://github.com/pypa/pip/issues/6334
  Building wheel for udf-ext (setup.py) ... error
  error: subprocess-exited-with-error

  × python setup.py bdist_wheel did not run successfully.
  │ exit code: 1
  ╰─> [107 lines of output]
      running bdist_wheel
      running build
      running build_ext
      building 'udf_ext' extension
      creating D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build
      creating D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310
      creating D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release
      creating D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release\src
      D:\Python\lib\site-packages\torch\utils\cpp_extension.py:2356: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation.
      If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST'].
        warnings.warn(
      Emitting ninja build file D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release\build.ninja...
      Compiling objects...
      Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
      [1/2] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin\nvcc --generate-dependencies-with-compile --dependency-output D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release\src/udf_kernel.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -ID:\Python\lib\site-packages\torch\include -ID:\Python\lib\site-packages\torch\include\torch\csrc\api\include "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\include" -ID:\Python\include -ID:\Python\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.43.34808\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.43.34808\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\cppwinrt" "-IC:\Program Files (x86)\Windows Kits\NETFXSDK\4.8\include\um" -c D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\src\udf_kernel.cu -o D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release\src/udf_kernel.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=udf_ext -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_89,code=compute_89 -gencode=arch=compute_89,code=sm_89
      FAILED: D:/Tests/Direct3D-S2/Direct3D-S2/third_party/voxelize/build/temp.win-amd64-cpython-310/Release/src/udf_kernel.obj
      C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin\nvcc --generate-dependencies-with-compile --dependency-output D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release\src/udf_kernel.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -ID:\Python\lib\site-packages\torch\include -ID:\Python\lib\site-packages\torch\include\torch\csrc\api\include "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\include" -ID:\Python\include -ID:\Python\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.43.34808\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.43.34808\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\cppwinrt" "-IC:\Program Files (x86)\Windows Kits\NETFXSDK\4.8\include\um" -c D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\src\udf_kernel.cu -o D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release\src/udf_kernel.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=udf_ext -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_89,code=compute_89 -gencode=arch=compute_89,code=sm_89
      D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\src\udf_kernel.cu(179): warning #2361-D: invalid narrowing conversion from "double" to "float"
                        Point3D queryPoint = {(float)i/(DIM-1) - 0.5, (float)j/(DIM-1) - 0.5, (float)k/(DIM-1) -0.5};
                                              ^

      Remark: The warnings can be suppressed with "-diag-suppress <warning-number>"

      D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\src\udf_kernel.cu(179): warning #2361-D: invalid narrowing conversion from "double" to "float"
                        Point3D queryPoint = {(float)i/(DIM-1) - 0.5, (float)j/(DIM-1) - 0.5, (float)k/(DIM-1) -0.5};
                                                                      ^

      D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\src\udf_kernel.cu(179): warning #2361-D: invalid narrowing conversion from "double" to "float"
                        Point3D queryPoint = {(float)i/(DIM-1) - 0.5, (float)j/(DIM-1) - 0.5, (float)k/(DIM-1) -0.5};
                                                                                              ^

      D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\src\udf_kernel.cu(182): error: expected a ")"
                  if (distance < threshold / DIM or  distance2 < threshold / DIM){

                                                 ^



      1 error detected in the compilation of "D:/Tests/Direct3D-S2/Direct3D-S2/third_party/voxelize/src/udf_kernel.cu".

      udf_kernel.cu

      [2/2] cl /showIncludes /nologo /O2 /W3 /GL /DNDEBUG /MD /MD /wd4819 /wd4251 /wd4244 /wd4267 /wd4275 /wd4018 /wd4190 /wd4624 /wd4067 /wd4068 /EHsc -ID:\Python\lib\site-packages\torch\include -ID:\Python\lib\site-packages\torch\include\torch\csrc\api\include "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\include" -ID:\Python\include -ID:\Python\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.43.34808\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.43.34808\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\cppwinrt" "-IC:\Program Files (x86)\Windows Kits\NETFXSDK\4.8\include\um" -c D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\src\udf_cuda.cpp /FoD:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\build\temp.win-amd64-cpython-310\Release\src/udf_cuda.obj -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=udf_ext -D_GLIBCXX_USE_CXX11_ABI=0 /std:c++17
      ninja: build stopped: subcommand failed.
      Traceback (most recent call last):
        File "D:\Python\lib\site-packages\torch\utils\cpp_extension.py", line 2506, in _run_ninja_build
          subprocess.run(
        File "D:\Python\lib\subprocess.py", line 526, in run
          raise CalledProcessError(retcode, process.args,
      subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

      The above exception was the direct cause of the following exception:

      Traceback (most recent call last):
        File "<string>", line 2, in <module>
        File "<pip-setuptools-caller>", line 35, in <module>
        File "D:\Tests\Direct3D-S2\Direct3D-S2\third_party\voxelize\setup.py", line 4, in <module>
          setup(
        File "D:\Python\lib\site-packages\setuptools\__init__.py", line 87, in setup
          return distutils.core.setup(**attrs)
        File "D:\Python\lib\site-packages\setuptools\_distutils\core.py", line 185, in setup
          return run_commands(dist)
        File "D:\Python\lib\site-packages\setuptools\_distutils\core.py", line 201, in run_commands
          dist.run_commands()
        File "D:\Python\lib\site-packages\setuptools\_distutils\dist.py", line 968, in run_commands
          self.run_command(cmd)
        File "D:\Python\lib\site-packages\setuptools\dist.py", line 1217, in run_command
          super().run_command(command)
        File "D:\Python\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command
          cmd_obj.run()
        File "D:\Python\lib\site-packages\wheel\_bdist_wheel.py", line 387, in run
          self.run_command("build")
        File "D:\Python\lib\site-packages\setuptools\_distutils\cmd.py", line 319, in run_command
          self.distribution.run_command(command)
        File "D:\Python\lib\site-packages\setuptools\dist.py", line 1217, in run_command
          super().run_command(command)
        File "D:\Python\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command
          cmd_obj.run()
        File "D:\Python\lib\site-packages\setuptools\_distutils\command\build.py", line 132, in run
          self.run_command(cmd_name)
        File "D:\Python\lib\site-packages\setuptools\_distutils\cmd.py", line 319, in run_command
          self.distribution.run_command(command)
        File "D:\Python\lib\site-packages\setuptools\dist.py", line 1217, in run_command
          super().run_command(command)
        File "D:\Python\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command
          cmd_obj.run()
        File "D:\Python\lib\site-packages\setuptools\command\build_ext.py", line 84, in run
          _build_ext.run(self)
        File "D:\Python\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 346, in run
          self.build_extensions()
        File "D:\Python\lib\site-packages\torch\utils\cpp_extension.py", line 1010, in build_extensions
          build_ext.build_extensions(self)
        File "D:\Python\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 466, in build_extensions
          self._build_extensions_serial()
        File "D:\Python\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 492, in _build_extensions_serial
          self.build_extension(ext)
        File "D:\Python\lib\site-packages\setuptools\command\build_ext.py", line 246, in build_extension
          _build_ext.build_extension(self, ext)
        File "D:\Python\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 547, in build_extension
          objects = self.compiler.compile(
        File "D:\Python\lib\site-packages\torch\utils\cpp_extension.py", line 978, in win_wrap_ninja_compile
          _write_ninja_file_and_compile_objects(
        File "D:\Python\lib\site-packages\torch\utils\cpp_extension.py", line 2159, in _write_ninja_file_and_compile_objects
          _run_ninja_build(
        File "D:\Python\lib\site-packages\torch\utils\cpp_extension.py", line 2522, in _run_ninja_build
          raise RuntimeError(message) from e
      RuntimeError: Error compiling objects for extension
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
  ERROR: Failed building wheel for udf-ext
  Running setup.py clean for udf-ext
Failed to build udf-ext
ERROR: Failed to build installable wheels for some pyproject.toml based projects (udf-ext)

If that can be supported under Windows and the VRAM is OK for single GPU then it should run under Windows.

[SoftologyPro]

Here is the install.bat I currently have. Save this as install.bat into an empty directory, command prompt into it and run install.bat to install under Windows. Expects python 3.10.x and CUDA 12.8 for 50xx support.

If anyone can help get the udf_ext compiled and installing it would be greatly appreciated.

@echo off

cd
echo *** %time% *** Deleting Direct3D-S2 directory if it exists
if exist Direct3D-S2\. rd /S /Q Direct3D-S2

echo *** %time% *** Cloning Direct3D-S2 repository
git clone https://github.com/DreamTechAI/Direct3D-S2
cd Direct3D-S2

echo *** %time% *** Creating venv
python -m venv venv
call venv\scripts\activate.bat

echo *** %time% *** Upgrading pip
python -m pip install -U pip
python -m pip install pip==24.0
pip install wheel==0.45.1
pip install setuptools==65.5.0
pip install ninja==1.11.1.4

echo *** %time% *** Removing triton from requirements.txt
type requirements.txt | findstr /v triton > stripped.txt
del requirements.txt
ren stripped.txt requirements.txt

echo *** %time% *** Removing flash-attn from requirements.txt
type requirements.txt | findstr /v flash-attn > stripped.txt
del requirements.txt
ren stripped.txt requirements.txt

echo *** %time% *** Removing torchsparse from requirements.txt
type requirements.txt | findstr /v torchsparse > stripped.txt
del requirements.txt
ren stripped.txt requirements.txt

echo *** %time% *** Installing torch
pip install torch==2.7.0+cu128 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
echo *** %time% *** Installing torchsparse
pip install https://data.pyg.org/whl/torch-2.7.0+cu126/torch_sparse-0.6.18+pt27cu126-cp310-cp310-win_amd64.whl
echo *** %time% *** Installing requirements.txt
pip install -r requirements.txt
echo *** %time% *** Installing .
pip install .
echo *** %time% *** Installing flash-attn
pip install https://huggingface.co/datasets/Softology-Pro/VoC/resolve/main/flash_attn-2.7.4.post1-cp310-cp310-win_amd64.whl

echo *** Installing other requirements
pip install trimesh==4.6.10

echo *** %time% *** Installing GPU Torch
pip uninstall -y torch
pip uninstall -y torch
pip uninstall -y torch
pip install torch==2.7.0+cu128 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128

pip install triton-windows

call venv\scripts\deactivate.bat
cd..

echo *** %time% *** Finished Direct3D-S2 install
echo.
echo *** %time% *** Scroll up and check for errors.  Do not assume it worked.
pause

[Dibucci]

I was also trying to get this setup on my windows PC, and ran into issues with installing dependencies

I'm using a venv, with python venv, so it wouldn't conflict with anything already on my system but idk if that's that issue as pip install cannot find anything for:
triton==3.1.0
flash-attn
and, idk how to even work with this:

git+https://github.com/mit-han-lab/torchsparse.git
third_party/voxelize/

as im not a coder, i just follow instructions and pick up some basic knowledge here and there from working with AI stuff since it almost all uses python.

here are what I get when even trying to install the two packages that I mentioned above

Collecting flash-attn
  Using cached flash_attn-2.7.4.post1.tar.gz (6.0 MB)
  Installing build dependencies ... done
  Getting requirements to build wheel ... error
  error: subprocess-exited-with-error

  × Getting requirements to build wheel did not run successfully.
  │ exit code: 1
  ╰─> [17 lines of output]
      Traceback (most recent call last):
        File "C:\Users\Owner\Documents\3d Test\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 389, in <module>
          main()
        File "C:\Users\Owner\Documents\3d Test\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 373, in main
          json_out["return_val"] = hook(**hook_input["kwargs"])
        File "C:\Users\Owner\Documents\3d Test\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 143, in get_requires_for_build_wheel
          return hook(config_settings)
        File "C:\Users\Owner\AppData\Local\Temp\pip-build-env-xqanqrmi\overlay\Lib\site-packages\setuptools\build_meta.py", line 331, in get_requires_for_build_wheel
          return self._get_build_requires(config_settings, requirements=[])
        File "C:\Users\Owner\AppData\Local\Temp\pip-build-env-xqanqrmi\overlay\Lib\site-packages\setuptools\build_meta.py", line 301, in _get_build_requires
          self.run_setup()
        File "C:\Users\Owner\AppData\Local\Temp\pip-build-env-xqanqrmi\overlay\Lib\site-packages\setuptools\build_meta.py", line 512, in run_setup
          super().run_setup(setup_script=setup_script)
        File "C:\Users\Owner\AppData\Local\Temp\pip-build-env-xqanqrmi\overlay\Lib\site-packages\setuptools\build_meta.py", line 317, in run_setup
          exec(code, locals())
        File "<string>", line 22, in <module>
      ModuleNotFoundError: No module named 'torch'
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.

(3d Test) C:\Users\Owner\Documents\3d Test\Direct3D-S2>pip install triton==3.1.0
ERROR: Could not find a version that satisfies the requirement triton==3.1.0 (from versions: none)
ERROR: No matching distribution found for triton==3.1.0

[Dibucci]

well, figured out im dumb with the git hub and the folder path, but still trying to install either one ends in a fail... honestly, only resaon even I was trying to get this running was so i don't have to wait behind 70+ people on the demo page xD

[Stoobs]

Apparently it's a code bug in the voxelize code:

The file .\Direct3D-S2\third_party\voxelize\src\udf_kernel.cu contains CUDA C++ code. In C++ and CUDA, the logical OR operator is '||', not 'or'. The compiler is encountering or and doesn't know what to do with it, leading to the "expected a ')'" error because it's trying to parse it differently.

Changing line 182 in udf_kernel.cu to if (distance < threshold / DIM || distance2 < threshold / DIM){ fixes the build error.

I've managed to get the requirements to install now, just needed to also update setup.py to use triton-windows rather than triton, and just build flash-attn from source: pip install flash-attn --no-build-isolation

[SNAKEIX]

yup no luck so far for windows , there many problems with installing this model in windows , i think they need rebuild the method to fit for windows and test it on windows environment, people are not programmers to try to fix codes and bugs, so hope they fix everything and repost the correct setup for windows + conda environment

[Deathdadev]

My pull request here fixes the udf error: #10
However I am still editing the code to get it working on windows, the main issue being torchsparce

[SNAKEIX]

it's not just udf error , also triton not working on windows

[Deathdadev]

instead of triton, install triton-windows

[Aikon404]

install the install.bat from SoftologyPro
then go to third_party\voxelize\src folder
open udf_kernel.cu with Editor
STRG + f
serch for "if (distance < threshold / DIM or distance2 < threshold / DIM){"
replace the line with "if (distance < threshold / DIM || distance2 < threshold / DIM){"
save the file
go to the main Folder and open cmd
type "call venv\Scripts\activate.bat"
press Enter
run These skipts
pip install -r requirements.txt
pip install gradio
pip install timm kornia

have fun

[Aikon404]

thx to Stoobs and SoftologyPro

[Stoobs]

If you get an error with installing flash-attn on windows, add a "MANIFEST.in" file with the following contents in the root folder:

include LICENSE
include CONTRIBUTORS
include CHANGELOG
include README.md

then rerun pip install flash-attn --no-build-isolation

You may also want to edit app.py and change the launch params to share=False, I really dislike people making share=True the default - sharing the app with the world should never be the default.

[Stoobs]

Okay, now stuck at the torchsparse issue - clearly there's a difference between the official torchsparse module and the one referenced in the code here.

The official version doesn't have the same features at all.

[Aikon404]

git Clone https://github.com/mit-han-lab/torchsparse.git
go in torchspares Folder
copy torchsparse
go to venv\Lib\site-packages Folder
paste torchsparse
pip install rootpath

I'm at this point now, it generates something but at the end there is still an error

[SNAKEIX]

this is not going to work like this , it's like trying to patch error after error , the developers should find all the problems for windows , after everything rebuild and tested on windows , repost how to install for windows in simple way in the main page

[umarmuzammil]

Has anyone tried running this in a docker environment on windows ? I couldnt build it in docker either, I also run into issues with torchparse on linux subsystem.

Can atleast the devs confirm the CUDA version and Python. So Far I have tried 12.1 and 11.8 but no success.