Can't install cupy==7.8.0
See original GitHub issueDescription
Chainer uses cupy == 7.8.0, I’m not sure if newer versions would work.
Option 1:
pip install cupy==7.8.0
CUDA_PATH : /cvmfs/soft.computecanada.ca/easybuild/software/2020/Core/cudacore/11.0.2 NVTOOLSEXT_PATH : (none) NVCC : (none) ROCM_HOME : (none) Modules: cuda : Yes (version 11000) cusolver : Yes cudnn : Yes (version 8003) nccl : No -> Include files not found: ['nccl.h'] -> Check your CFLAGS environment variable. nvtx : Yes thrust : Yes cutensor : No -> Include files not found: ['cutensor.h'] -> Check your CFLAGS environment variable. cub : Yes WARNING: Some modules could not be configured. CuPy will be installed without these modules. Please refer to the Installation Guide for details: https://docs.cupy.dev/en/stable/install.html ************************************************************ NOTICE: Skipping cythonize as cupy/core/_dtype.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_kernel.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_memory_range.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_routines_indexing.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_routines_logic.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_routines_manipulation.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_routines_math.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_routines_sorting.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_routines_statistics.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/_scalar.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/core.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/dlpack.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/flags.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/internal.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/fusion.pyx does not exist. NOTICE: Skipping cythonize as cupy/core/raw.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/cublas.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/cufft.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/curand.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/cusparse.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/device.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/driver.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/memory.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/memory_hook.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/nvrtc.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/pinned_memory.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/profiler.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/function.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/stream.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/runtime.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/texture.pyx does not exist. NOTICE: Skipping cythonize as cupy/util.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/cusolver.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/cudnn.pyx does not exist. NOTICE: Skipping cythonize as cupy/cudnn.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/nvtx.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/thrust.pyx does not exist. NOTICE: Skipping cythonize as cupy/cuda/cub.pyx does not exist. Traceback (most recent call last): File "<string>", line 2, in <module> File "<pip-setuptools-caller>", line 34, in <module> File "/tmp/pip-install-3nvzj2jg/cupy_b0144fd6a6cf445885b7f609ca09f2cc/setup.py", line 156, in <module> setup( File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/__init__.py", line 87, in setup return distutils.core.setup(**attrs) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/core.py", line 177, in setup return run_commands(dist) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/core.py", line 193, in run_commands dist.run_commands() File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/dist.py", line 968, in run_commands self.run_command(cmd) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/dist.py", line 1217, in run_command super().run_command(command) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/dist.py", line 987, in run_command cmd_obj.run() File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/command/install.py", line 68, in run return orig.install.run(self) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/command/install.py", line 695, in run self.run_command('build') File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/cmd.py", line 317, in run_command self.distribution.run_command(command) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/dist.py", line 1217, in run_command super().run_command(command) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/dist.py", line 987, in run_command cmd_obj.run() File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/command/build.py", line 24, in run super().run() File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/command/build.py", line 131, in run self.run_command(cmd_name) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/cmd.py", line 317, in run_command self.distribution.run_command(command) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/dist.py", line 1217, in run_command super().run_command(command) File "/home/jolicoea/vidgen2/lib/python3.8/site-packages/setuptools/_distutils/dist.py", line 987, in run_command cmd_obj.run() File "/tmp/pip-install-3nvzj2jg/cupy_b0144fd6a6cf445885b7f609ca09f2cc/cupy_setup_build.py", line 991, in run check_extensions(self.extensions) File "/tmp/pip-install-3nvzj2jg/cupy_b0144fd6a6cf445885b7f609ca09f2cc/cupy_setup_build.py", line 728, in check_extensions raise RuntimeError('''\ RuntimeError: Missing file: cupy/cuda/cub.cpp Please install Cython 0.28.0 or later. Please also check the version of Cython. See https://docs.cupy.dev/en/stable/install.html for details.
I have Cython 0.29.31, pip install says that 0.28.0 is not available so I can’t downgrade.
Option 2:
pip install cupy
or install from a git clone with pip install -e .
Install properly, but chainer does not think that cupy is properly installed and I get cupy.core not found errors
Option 3:
pip install cupy-cuda110
ERROR: Could not find a version that satisfies the requirement cupy-cuda110 (from versions: none)
ERROR: No matching distribution found for cupy-cuda110
I think using version 7.8.0 of cupy would make it work with chainer, but I can’t install it.
To Reproduce
pip install cupy==7.8.0 # error
pip install cupy-cuda110 # error not found
pip install cupy # works
pip install chainer
python -c 'import chainer; chainer.print_runtime_info()' # cupy not recognized by chainer
Installation
Source (pip install cupy
)
Environment
OS : Linux-3.10.0-1160.62.1.el7.x86_64-x86_64-with-glibc2.2.5
Python Version : 3.8.10
CuPy Version : 11.0.0
CuPy Platform : NVIDIA CUDA
NumPy Version : 1.23.0
SciPy Version : 1.8.0
Cython Build Version : 0.29.31
Cython Runtime Version : 0.29.31
CUDA Root : /cvmfs/soft.computecanada.ca/easybuild/software/2020/Core/cudacore/11.0.2
nvcc PATH : /cvmfs/soft.computecanada.ca/easybuild/software/2020/Core/cudacore/11.0.2/bin/nvcc
CUDA Build Version : 11040
CUDA Driver Version : 11060
CUDA Runtime Version : 11040
cuBLAS Version : (available)
cuFFT Version : 10502
cuRAND Version : 10205
cuSOLVER Version : (11, 2, 0)
cuSPARSE Version : (available)
NVRTC Version : (11, 4)
Thrust Version : 101201
CUB Build Version : 101201
Jitify Build Version : d90e2e0
cuDNN Build Version : 8200
cuDNN Version : 8200
NCCL Build Version : 21104
NCCL Runtime Version : 21104
cuTENSOR Version : None
cuSPARSELt Build Version : None
Device 0 Name : Tesla V100-SXM2-16GB
Device 0 Compute Capability : 70
Device 0 PCI Bus ID : 0000:1C:00.0
Device 1 Name : Tesla V100-SXM2-16GB
Device 1 Compute Capability : 70
Device 1 PCI Bus ID : 0000:1D:00.0
Additional Information
No response
Issue Analytics
- State:
- Created a year ago
- Comments:12 (7 by maintainers)
Top Results From Across the Web
Installation — CuPy 11.4.0 documentation
SciPy and Optuna are optional dependencies and will not be installed automatically. Note. Before installing CuPy, we recommend you to upgrade setuptools and...
Read more >Can't install via Pip · Issue #952 · cupy/cupy - GitHub
Tried to install via pip install cupy but got the following error: Collecting cupy Using cached cupy-2.3.0.tar.gz Complete output from ...
Read more >I cannot install cupy with pip - Stack Overflow
I have tried updating setuptools and purging and reinstalling pip. I am running Linux Mint 19.3 Cinnamon 4.4.8. If anyone has experienced this ......
Read more >Habitat Threats in the Sagebrush Ecosystem: | USDA Forest Service
Chapter 8: Utility of Greater Sage-Grouse as an Umbrella Species ... cant ramifications not only from an eco- ... tation, and connectivity)-can be...
Read more >Installation — Chainer 7.8.1 documentation
You can accelerate performance of Chainer by installing the following optional components. ... CuPy v7.8.0 is the recommended version for Chainer v7 series....
Read more >Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start FreeTop Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found
Top GitHub Comments
v7.8.0 sdist was broken and we have a hotfix as v7.8.0.post1.
https://github.com/cupy/cupy/releases/v7.8.0.post1 https://pypi.org/project/cupy/#history
So please use
pip install cupy==v7.8.0.post1
in case you need CuPy v7.8.0.thanks @leofang, I’ll reach out to them!