When doing pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./, shows ModuleNotFoundError: No module named 'packaging'

[lainmn]

Describe the Bug

When doing pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./, shows ModuleNotFoundError: No module named 'packaging', but I have packaging installed

The detailed error looks like this: Traceback (most recent call last): File "C:\Users\luo00318\AppData\Roaming\Python\Python39\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 353, in main() File "C:\Users\luo00318\AppData\Roaming\Python\Python39\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 335, in main json_out['return_val'] = hook(**hook_input['kwargs']) File "C:\Users\luo00318\AppData\Roaming\Python\Python39\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 118, in get_requires_for_build_wheel return hook(config_settings) File "C:\Users\luo00318\AppData\Local\Temp\pip-build-env-wchsm8v0\overlay\Lib\site-packages\setuptools\build_meta.py", line 355, in get_requires_for_build_wheel return self._get_build_requires(config_settings, requirements=['wheel']) File "C:\Users\luo00318\AppData\Local\Temp\pip-build-env-wchsm8v0\overlay\Lib\site-packages\setuptools\build_meta.py", line 325, in _get_build_requires self.run_setup() File "C:\Users\luo00318\AppData\Local\Temp\pip-build-env-wchsm8v0\overlay\Lib\site-packages\setuptools\build_meta.py", line 341, in run_setup exec(code, locals()) File "", line 5, in ModuleNotFoundError: No module named 'packaging' 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. full command: 'C:\Users\luo00318\AppData\Local\anaconda3\envs\torch_gpu\python.exe' 'C:\Users\luo00318\AppData\Roaming\Python\Python39\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py' get_requires_for_build_wheel 'C:\Users\luo00318\AppData\Local\Temp\tmptlqivujq' cwd: C:\Users\luo00318\Research\ICL\myplan\week4\apex Getting requirements to build wheel ... error 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.

Minimal Steps/Code to Reproduce the Bug

Expected Behavior

should have been installed correctly?

Environment

[orrzohar]

I got the same error.

[sym330]

I got the same error.

[lengendhao]

Hey! I also got this error, and I've solved it. You can see #1653, which said:

1. clone or download apex 22.04-dev branch instead of master brach
2. use the command：pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./ to install 'apex'. You may get another error: No such file or directory: ':/usr/local/cuda-11.0/bin/nvcc' You can vi ~/.bashrc then change 'export CUDA_HOME=$CUDA_HOME:/usr/local/cuda' into 'export CUDA_HOME=/usr/local/cuda' Finally, source ~/.bashrc Use the command again, and you will use apex succesfully!

[rocke2020]

Hey! I also got this error, and I've solved it. You can see #1653, which said:

1. clone or download apex 22.04-dev branch instead of master brach
2. use the command：pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./ to install 'apex'. You may get another error: No such file or directory: ':/usr/local/cuda-11.0/bin/nvcc' You can vi ~/.bashrc then change 'export CUDA_HOME=$CUDA_HOME:/usr/local/cuda' into 'export CUDA_HOME=/usr/local/cuda' Finally, source ~/.bashrc Use the command again, and you will use apex succesfully!

could you share your package details? thanks in advance! 1, torch version 2, python version 3, cuda version My envs: torch 2.1, py 3.10, cuda 11.8, I has the same apex install error. the 22.04-dev branch also failed to install.

[lengendhao]

Hey! I also got this error, and I've solved it. You can see #1653, which said:

1. clone or download apex 22.04-dev branch instead of master brach
2. use the command：pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./ to install 'apex'. You may get another error: No such file or directory: ':/usr/local/cuda-11.0/bin/nvcc' You can vi ~/.bashrc then change 'export CUDA_HOME=$CUDA_HOME:/usr/local/cuda' into 'export CUDA_HOME=/usr/local/cuda' Finally, source ~/.bashrc Use the command again, and you will use apex succesfully!

could you share your package details? thanks in advance! 1, torch version 2, python version 3, cuda version My envs: torch 2.1, py 3.10, cuda 11.8, I has the same apex install error. the 22.04-dev branch also failed to install.

My envs: PyTorch 1.8.1 Python 3.8(ubuntu18.04) Cuda 11.1 Maybe this issue can help you: https://github.com/NVIDIA/apex/issues/1735

[rocke2020]

Hey! I also got this error, and I've solved it. You can see #1653, which said:

1. clone or download apex 22.04-dev branch instead of master brach
2. use the command：pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./ to install 'apex'. You may get another error: No such file or directory: ':/usr/local/cuda-11.0/bin/nvcc' You can vi ~/.bashrc then change 'export CUDA_HOME=$CUDA_HOME:/usr/local/cuda' into 'export CUDA_HOME=/usr/local/cuda' Finally, source ~/.bashrc Use the command again, and you will use apex succesfully!

could you share your package details? thanks in advance! 1, torch version 2, python version 3, cuda version My envs: torch 2.1, py 3.10, cuda 11.8, I has the same apex install error. the 22.04-dev branch also failed to install.

My envs: PyTorch 1.8.1 Python 3.8(ubuntu18.04) Cuda 11.1 Maybe this issue can help you: #1735

@lengendhao thanks!! I create an env exactly as yours, and with 22.04-dev, finally apex successfully installed with pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./

[zhongyu-zhao]

python 3.7.3 pytorch 1.5.1 cuda 12.0 conda install packaging pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --global-option="--cpp_ext" --global-option="--cuda_ext" ./ apex 22.04-dev branch doesn't work for me... but i've found that #1748 solved my problem. Good luck!

[LukeLIN-web]

I also met this problem. https://github.com/NVIDIA/apex/issues/1748#issuecomment-1928910265 solved my problem.