ERROR when pip install -e . in windows 11, RTX 5070

[HkDzl]

(3dgrut) PS D:\CodeBase\3DReconstruction\3dgrut\thirdparty\fused-ssim> pip install -e .
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
Obtaining file:///D:/CodeBase/3DReconstruction/3dgrut/thirdparty/fused-ssim
  Installing build dependencies ... done
  Checking if build backend supports build_editable ... done
  Getting requirements to build editable ... error
  error: subprocess-exited-with-error

  × Getting requirements to build editable did not run successfully.
  │ exit code: 1
  ╰─> [35 lines of output]
      C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\torch\_subclasses\functional_tensor.py:283: UserWarning: Failed to initialize NumPy: No module named 'numpy' (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\pytorch\torch\csrc\utils\tensor_numpy.cpp:84.)
        cpu = _conversion_method_template(device=torch.device("cpu"))
      Compiling for CUDA.
      Failed to detect GPU architecture: Torch not compiled with CUDA enabled. Falling back to multiple architectures.
      Compiling for CUDA.
      Failed to detect GPU architecture: Torch not compiled with CUDA enabled. Falling back to multiple architectures.
      Traceback (most recent call last):
        File "D:\ProgramData\miniconda3\envs\3dgrut\Lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 389, in <module>
          main()
        File "D:\ProgramData\miniconda3\envs\3dgrut\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 "D:\ProgramData\miniconda3\envs\3dgrut\Lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 157, in get_requires_for_build_editable
          return hook(config_settings)
                 ^^^^^^^^^^^^^^^^^^^^^
        File "C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\setuptools\build_meta.py", line 479, in get_requires_for_build_editable
          return self.get_requires_for_build_wheel(config_settings)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\setuptools\build_meta.py", line 333, in get_requires_for_build_wheel
          return self._get_build_requires(config_settings, requirements=[])
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\setuptools\build_meta.py", line 301, in _get_build_requires
          self.run_setup()
        File "C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\setuptools\build_meta.py", line 317, in run_setup
          exec(code, locals())
        File "<string>", line 128, in <module>
        File "C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\torch\utils\cpp_extension.py", line 1408, in CUDAExtension
          library_dirs += library_paths(device_type="cuda")
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\torch\utils\cpp_extension.py", line 1641, in library_paths
          paths.append(_join_cuda_home(lib_dir))
                       ^^^^^^^^^^^^^^^^^^^^^^^^
        File "C:\Users\admin\AppData\Local\Temp\pip-build-env-k0lp6yeg\overlay\Lib\site-packages\torch\utils\cpp_extension.py", line 3102, in _join_cuda_home
          raise OSError('CUDA_HOME environment variable is not set. '
      OSError: CUDA_HOME environment variable is not set. Please set it to your CUDA install root.
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed to build 'file:///D:/CodeBase/3DReconstruction/3dgrut/thirdparty/fused-ssim' when getting requirements to build editable

My system is Windows 11 and my GPU is 5070. Has anyone encountered a similar problem?(CUDA 12.8 and torch version=2.10+cu128)

[rahul-goel]

If it is of any help, I think you have to set the CUDA_HOME environment variable to the correct directory path of your native CUDA installation. If you're using conda, it is going to be the same as CONDA_PREFIX.

I don't have any experience in Windows: this is just a guess. I hope it helps.