nvdiffrec Error in installation

run this line: pip install --global-option="--no-networks" git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch

then get this error : ERROR: Command errored out with exit status 1: command: 'C:\Users\ReZaN.conda\envs\dmodel\python.exe' -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] ="'"'C:\Users\ReZaN\AppData\Local\Temp\pip-req-build-9qo96lf9\bindings/torch\setup.py'"'"'; file='"'"'C:\Users\ReZaN\AppData\Local\Temp\pip-req-build-9qo96lf9\bindings/torch\setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(file) if os.path.exists(file) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' egg_info --egg-base 'C:\Users\ReZaN\AppData\Local\Temp\pip-pip-egg-info-gryf3c1b' cwd: C:\Users\ReZaN\AppData\Local\Temp\pip-req-build-9qo96lf9\bindings/torch Complete output (6 lines): Traceback (most recent call last): File "", line 1, in File "C:\Users\ReZaN\AppData\Local\Temp\pip-req-build-9qo96lf9\bindings/torch\setup.py", line 41, in raise RuntimeError("Could not locate a supported Microsoft Visual C++ installation") RuntimeError: Could not locate a supported Microsoft Visual C++ installation Building PyTorch extension for tiny-cuda-nn version 1.6 ---------------------------------------- WARNING: Discarding git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch. Command errored out with exit status 1: python setup.py egg_info Check the logs for full command output. ERROR: Command errored out with exit status 1: python setup.py egg_info Check the logs for full command output.

May 23 '22 06:05 rezanit76

As noted in our installation instructions, we require Visual Studio to build tiny-cuda-nn. From the log above, it seems VS is not found in your env.

raise RuntimeError("Could not locate a supported Microsoft Visual C++ installation")
RuntimeError: Could not locate a supported Microsoft Visual C++ installation

If the error persists, I would suggest reporting that issue there https://github.com/NVlabs/tiny-cuda-nn, or testing their bindings in isolation to track down the issue: https://github.com/NVlabs/tiny-cuda-nn#pytorch-extension

May 23 '22 06:05 jmunkberg

@rezanit76 you need to install the vs_build_tools download, which will setup a supported Microsoft Visual C++ installation. That should be enough to fix that error. That's how I fixed it.

May 24 '22 02:05 ECE-Engineer

I´ve got the same error. In the TinyCuda documentations they suggest to install Visual Studio Code, C++ complements, and at least run "helloWorld". However, I kept getting the error: "Could not locate a supported Microsoft Visual C++ installation".

What worked for me instantly was to add in PATH the location of "cl.exe". It is embedded in the installation of Microsoft Visual Studio (probably Visual Studio Code is not a must, but I'm not sure). In Windows, go to System Properties >> Advanced system settings >> Advanced (tab) >> Environment variables >> PATH (edit) >> here I've added a new line with: C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.32.31326\bin\Hostx64\x64 This is the location of "cl.exe" which is the C++ compiler.

I've had to add several variables in PATH. Just for the record this is my entire PATH: C:\Program Files\Git\bin;C:\Program Files\Git\cmd;C:\Users\jpyunda\Documents\Nvidia CUDA docs\zlib123dllx64;C:\Program Files\NVIDIA\CUDNN\v 8.4.1.50;C:\Program Files\CMake\bin;C:\Users\jpyunda\AppData\Local\Programs\Microsoft VS Code\bin;C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.32.31326\bin\Hostx64\x64;