I'm trying to install cuDF to gain the glorious powers of multithreading my NVIDIA GPU in my python scripts. However, I can't install cuDF.
To install cuDF, i went to https://docs.rapids.ai/install and ran the created command:
pip install \ --extra-index-url=https://pypi.nvidia.com \ cudf-cu12==24.6.* dask-cudf-cu12==24.6.* cuml-cu12==24.6.* \ cugraph-cu12==24.6.* cuspatial-cu12==24.6.* cuproj-cu12==24.6.* \ cuxfilter-cu12==24.6.* cucim-cu12==24.6.* pylibraft-cu12==24.6.* \ raft-dask-cu12==24.6.* cuvs-cu12==24.6.*
but it produced an error in which it says "Didn't find wheel for cudf-cu12 24.6.1" (full error message below). In the error message, it even suggests to run the command
$ pip install --extra-index-url https://pypi.nvidia.com cudf-cu12
but that produces the same error message- ironically with that same suggestion within it.
How can I proceed? I installed the CUDA 12 driver, I downgraded my Python to Python 11 as required, I tried the conda install, I tried installing lower versions (CUDA 11 instead of CUDA 12).
Here is the full output from running the pip install:
%pip install \
--extra-index-url=https://pypi.nvidia.com \
cudf-cu12==24.6.* dask-cudf-cu12==24.6.* cuml-cu12==24.6.* \
cugraph-cu12==24.6.* cuspatial-cu12==24.6.* cuproj-cu12==24.6.* \
cuxfilter-cu12==24.6.* cucim-cu12==24.6.* pylibraft-cu12==24.6.* \
raft-dask-cu12==24.6.* cuvs-cu12==24.6.*
Looking in indexes: https://pypi.org/simple, https://pypi.nvidia.com Collecting cudf-cu12==24.6.* Using cached cudf_cu12-24.6.1.tar.gz (2.6 kB) Installing build dependencies: started Installing build dependencies: finished with status 'done' Getting requirements to build wheel: started Getting requirements to build wheel: finished with status 'done' Preparing metadata (pyproject.toml): started
Preparing metadata (pyproject.toml): finished with status 'error' Note: you may need to restart the kernel to use updated packages.
error: subprocess-exited-with-error Preparing metadata (pyproject.toml) did not run successfully. exit code: 1 [62 lines of output] INFO:nvidia-stub:Testing wheel cudf_cu12-24.6.1-cp310-cp310-manylinux_2_28_aarch64.whl against tag cp310-cp310-manylinux_2_28_aarch64 INFO:nvidia-stub:Testing wheel cudf_cu12-24.6.1-cp310-cp310-manylinux_2_28_x86_64.whl against tag cp310-cp310-manylinux_2_28_x86_64 INFO:nvidia-stub:Testing wheel cudf_cu12-24.6.1-cp311-cp311-manylinux_2_28_aarch64.whl against tag cp311-cp311-manylinux_2_28_aarch64 INFO:nvidia-stub:Testing wheel cudf_cu12-24.6.1-cp311-cp311-manylinux_2_28_x86_64.whl against tag cp311-cp311-manylinux_2_28_x86_64 INFO:nvidia-stub:Testing wheel cudf_cu12-24.6.1-cp39-cp39-manylinux_2_28_aarch64.whl against tag cp39-cp39-manylinux_2_28_aarch64 INFO:nvidia-stub:Testing wheel cudf_cu12-24.6.1-cp39-cp39-manylinux_2_28_x86_64.whl against tag cp39-cp39-manylinux_2_28_x86_64 File "C:\Users\samt\AppData\Local\Temp\pip-build-env-7iipheuq\overlay\Lib\site-packages\nvidia_stub\wheel.py", line 177, in download_wheel return download_manual(wheel_directory, distribution, version)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\samt\AppData\Local\Temp\pip-build-env-7iipheuq\overlay\Lib\site-packages\nvidia_stub\wheel.py", line 144, in download_manual raise RuntimeError(f"Didn't find wheel for {distribution} {version}") Traceback (most recent call last): File "C:\Users\samt\AppData\Local\Temp\pip-build-env-7iipheuq\overlay\Lib\site-packages\nvidia_stub\wheel.py", line 177, in download_wheel return download_manual(wheel_directory, distribution, version)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\samt\AppData\Local\Temp\pip-build-env-7iipheuq\overlay\Lib\site-packages\nvidia_stub\wheel.py", line 144, in download_manual raise RuntimeError(f"Didn't find wheel for {distribution} {version}") RuntimeError: Didn't find wheel for cudf-cu12 24.6.1 During handling of the above exception, another exception occurred: Traceback (most recent call last):
File "C:\Users\samt\anaconda3\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 353, in main() File "C:\Users\samt\anaconda3\Lib\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\samt\anaconda3\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 152, in prepare_metadata_for_build_wheel whl_basename = backend.build_wheel(metadata_directory, config_settings)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\samt\AppData\Local\Temp\pip-build-env-7iipheuq\overlay\Lib\site-packages\nvidia_stub\buildapi.py", line 29, in build_wheel return download_wheel(pathlib.Path(wheel_directory), config_settings)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\samt\AppData\Local\Temp\pip-build-env-7iipheuq\overlay\Lib\site-packages\nvidia_stub\wheel.py", line 179, in download_wheel report_install_failure(distribution, version, exception_context) File "C:\Users\samt\AppData\Local\Temp\pip-build-env-7iipheuq\overlay\Lib\site-packages\nvidia_stub\error.py", line 63, in report_install_failure raise InstallFailedError(
nvidia_stub.error.InstallFailedError:
******************************************************************************* The installation of cudf-cu12 for version 24.6.1 failed. This is a special placeholder package which downloads a real wheel package
from https://pypi.nvidia.com. If https://pypi.nvidia.com is not reachable, we cannot download the real wheel file to install.
You might try installing this package via$ pip install --extra-index-url https://pypi.nvidia.com cudf-cu12
Here is some debug information about your platform to include in any bug report: Python Version: CPython 3.11.9 Operating System: Windows 10 CPU Architecture: AMD64 Driver Version: 555.99 CUDA Version: 12.5 ******************************************************************************* [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. error: metadata-generation-failed Encountered error while generating package metadata. See above for output. note: This is an issue with the package mentioned above, not pip. hint: See above for details.
I also tried the conda version of the install created from the RAPIDS site:
conda create -n rapids-24.06 -c rapidsai -c conda-forge -c nvidia rapids=24.06 python=3.11 'cuda-version>=12.0,<=12.2'
and that produced:
Channels:
- rapidsai
- conda-forge
- nvidia
- defaults
Platform: win-64
Collecting package metadata (repodata.json): done
Solving environment: failed
PackagesNotFoundError: The following packages are not available from current channels:
- rapids=24.06*
Current channels:
- https://conda.anaconda.org/rapidsai
- https://conda.anaconda.org/conda-forge
- https://conda.anaconda.org/nvidia
- defaults
To search for alternate channels that may provide the conda package you're
looking for, navigate to
https://anaconda.org
I just don't know how to proceed. Maybe I need to research how to install wheels?
UPDATE: The error message states I am using microsoft 10, but according to my computer, I am not:
Edition Windows 11 Pro Version 22H2 Installed on 2/4/2023 OS build 22621.3958 Experience Windows Feature Experience Pack 1000.22700.1026.0
UPDATE: I am now using the WSL2 terminal. I installed anaconda, and I installed RAPIDS as instructed on the website in the WS2L:
conda create -n rapids-24.06 -c rapidsai -c conda-forge -c nvidia \
rapids=24.06 python=3.11 'cuda-version>=12.0,<=12.2
And although this was successful, there are still problems. Even inside the WSL2, I cannot import cudf; it gives the module not found error
(base) aridleprechaun@Samuel-Tanner-Desktop:~$ ls
Anaconda3-2024.06-1-Linux-x86_64.sh Anaconda3-2024.06-1-Windows-x86_64.exe anaconda3
Anaconda3-2024.06-1-Linux-x86_64.sh.1 Anaconda3-2024.06-1-Windows-x86_64.exe.1
(base) aridleprechaun@Samuel-Tanner-Desktop:~$ python
Python 3.12.4 | packaged by Anaconda, Inc. | (main, Jun 18 2024, 15:12:24) [GCC 11.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import cudf
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ModuleNotFoundError: No module named 'cudf'
cudf should be already be installed when I installed RAPIDS. Proof/verification I installed RAPIDS:
(base) aridleprechaun@Samuel-Tanner-Desktop:~$ conda create -n rapids-24.06 -c rapidsai -c conda-forge -c nvidia \
rapids=24.06 python=3.11 'cuda-version>=12.0,<=12.2'
WARNING: A conda environment already exists at '/home/aridleprechaun/anaconda3/envs/rapids-24.06'
Remove existing environment (y/[n])? n
CondaSystemExit: Exiting.
It looks like you're installing on Windows. To use cuDF on windows, you need to use Windows Subsystem for Linux. The RAPIDS install page has a section for installing on WSL that should hopefully get you up and running.