conda install packaging

not works, the bug still exist

try add --no-build-isolation flag in pip install.

Doesn’t work for me, packaging already installed

Using the release 23.05 works for me (https://github.com/NVIDIA/apex/tags) Hash tag: 0da3ffb

try add --no-build-isolation flag in pip install.

It works for me. The full command is: pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings "--build-option=--cpp_ext" --config-settings "--build-option=--cuda_ext" ./

try add --no-build-isolation flag in pip install.

After trying all the solutions above (but the issue still kept unresolved), this one eventually worked for me 👍

same problem， Using pip 22.3.1 from /home/panz/anaconda3/envs/gpt/lib/python3.8/site-packages/pip (python 3.8) WARNING: Implying --no-binary=:all: due to the presence of --build-option / --global-option / --install-option. Consider using --config-settings for more flexibility. DEPRECATION: --no-binary currently disables reading from the cache of locally built wheels. In the future --no-binary will not influence the wheel cache. pip 23.1 will enforce this behaviour change. A possible replacement is to use the --no-cache-dir option. You can use the flag --use-feature=no-binary-enable-wheel-cache to test the upcoming behaviour. Discussion can be found at https://github.com/pypa/pip/issues/11453 Processing /home/panz/project/ColossalAI/examples/language/gpt/gemini/apex Running command python setup.py egg_info

torch.version = 1.12.0+cu113

running egg_info creating /tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info writing /tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info/PKG-INFO writing dependency_links to /tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info/dependency_links.txt writing requirements to /tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info/requires.txt writing top-level names to /tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info/top_level.txt writing manifest file ‘/tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info/SOURCES.txt’ reading manifest file ‘/tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info/SOURCES.txt’ adding license file ‘LICENSE’ writing manifest file ‘/tmp/pip-pip-egg-info-7w8pc_qc/apex.egg-info/SOURCES.txt’ Preparing metadata (setup.py) … done Requirement already satisfied: packaging>20.6 in /home/panz/anaconda3/envs/gpt/lib/python3.8/site-packages (from apex==0.1) (23.0) Installing collected packages: apex DEPRECATION: apex is being installed using the legacy ‘setup.py install’ method, because the ‘–no-binary’ option was enabled for it and this currently disables local wheel building for projects that don’t have a ‘pyproject.toml’ file. pip 23.1 will enforce this behaviour change. A possible replacement is to enable the ‘–use-pep517’ option. Discussion can be found at https://github.com/pypa/pip/issues/11451 Running command Running setup.py install for apex

torch.version = 1.12.0+cu113

Compiling cuda extensions with nvcc: NVIDIA ® Cuda compiler driver Copyright © 2005-2023 NVIDIA Corporation Built on Tue_Feb__7_19:32:13_PST_2023 Cuda compilation tools, release 12.1, V12.1.66 Build cuda_12.1.r12.1/compiler.32415258_0 from /home/panz/anaconda3/envs/gpt/bin

Traceback (most recent call last): File “<string>”, line 2, in <module> File “<pip-setuptools-caller>”, line 34, in <module> File “/home/panz/project/ColossalAI/examples/language/gpt/gemini/apex/setup.py”, line 171, in <module> check_cuda_torch_binary_vs_bare_metal(CUDA_HOME) File “/home/panz/project/ColossalAI/examples/language/gpt/gemini/apex/setup.py”, line 33, in check_cuda_torch_binary_vs_bare_metal raise RuntimeError( RuntimeError: Cuda extensions are being compiled with a version of Cuda that does not match the version used to compile Pytorch binaries. Pytorch binaries were compiled with Cuda 11.3. In some cases, a minor-version mismatch will not cause later errors: https://github.com/NVIDIA/apex/pull/323#discussion_r287021798. You can try commenting out this check (at your own risk). error: subprocess-exited-with-error

× Running setup.py install for apex 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: /home/panz/anaconda3/envs/gpt/bin/python -u -c ’ exec(compile(‘"’“‘’”‘"’‘"’"’

This is <pip-setuptools-caller> – a caller that pip uses to run setup.py

- It imports setuptools before invoking setup.py, to enable projects that directly

import from distutils.core to work with newer packaging standards.

- It provides a clear error message when setuptools is not installed.

- It sets sys.argv[0] to the underlying setup.py, when invoking setup.py so

setuptools doesn’“'”'t think the script is -c. This avoids the following warning:

manifest_maker: standard file ‘"’“‘-c’”‘"’ not found".

- It generates a shim setup.py, for handling setup.cfg-only projects.

import os, sys, tokenize

try: import setuptools except ImportError as error: print( "ERROR: Can not execute setup.py since setuptools is not available in " “the build environment.”, file=sys.stderr, ) sys.exit(1)

file = %r sys.argv[0] = file

if os.path.exists(file): filename = file with tokenize.open(file) as f: setup_py_code = f.read() else: filename = “<auto-generated setuptools caller>” setup_py_code = “from setuptools import setup; setup()”

exec(compile(setup_py_code, filename, “exec”)) ‘"’“‘’”‘"’‘"’“’ % ('”‘"’/home/panz/project/ColossalAI/examples/language/gpt/gemini/apex/setup.py’“'”‘,), “<pip-setuptools-caller>”, “exec”))’ --cpp_ext --cuda_ext install --record /tmp/pip-record-ss3s_tnl/install-record.txt --single-version-externally-managed --compile --install-headers /home/panz/anaconda3/envs/gpt/include/python3.8/apex cwd: /home/panz/project/ColossalAI/examples/language/gpt/gemini/apex/ Running setup.py install for apex … error error: legacy-install-failure

× Encountered error while trying to install package. ╰─> apex

note: This is an issue with the package mentioned above, not pip. hint: See above for output from the failure.

pip uninstall setuptools

pip install setuptools==60.2.0

pip install packaging

rm -R apex

git clone https://github.com/NVIDIA/apex

cd apex

pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings “–build-option=–cpp_ext” --config-settings “–build-option=–cuda_ext” ./

~/It will work 😃

conda install packaging not works I tried to ‘git checkout’ a older version and apex was installed sucessfully

Which older version did you use?

I checkout the commit “6943fd26e04c59327de32592cf5af68be8f5c44e” and it works (actually chosen randomly an older one)

pip uninstall setuptools

pip install setuptools==60.2.0

pip install packaging

rm -R apex

git clone https://github.com/NVIDIA/apex

cd apex

pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings “–build-option=–cpp_ext” --config-settings “–build-option=–cuda_ext” ./

~/It will work 😃

it did! ♥️

cuda 11.7 python3.7.6 torch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1

unzip apex.zip cd apex

1、install no cuda python setup.py install

2、 install with cuda python setup.py install --cuda_ext The compilation process is very long, at least ten minutes, and the CPU and memory footprint is 8c 8G

Doesn’t work for me, packaging already installed

I came to the same situation. I have solved it by installing the packaging by conda install packaging rather than pip install packaging.

And APEX gets installed successfully for me.

File “/home/glm/apex/setup.py”, line 4, in from packaging.version import parse, Version ModuleNotFoundError: No module named ‘packaging’

could you install packaging and retry?

pip uninstall setuptools

pip install setuptools==60.2.0

pip install packaging

rm -R apex

git clone https://github.com/NVIDIA/apex

cd apex

pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings “–build-option=–cpp_ext” --config-settings “–build-option=–cuda_ext” ./

~/It will work 😃

it works! thanks mikenetrino 😃

6943fd26e04c59327de32592cf5af68be8f5c44e

I chosed 22.3 in tags，successed