1

I need help. I am using Google Colab with Python 3.10.11 and I have a Colab with CUDA Version: 12.0 , Nvidia driver version 525.85.12 and I am following this tutorial on how to install cuDF https://colab.research.google.com/drive/1TAAi_szMfWqRfHVfjGSqnGVLr_ztzUM9#scrollTo=G_2dancFhf70

I checked
nvidia-smi

and I have been allocated Tesla T4

On the step
!python rapidsai-csp-utils/colab/install_rapids.py stable
import os
os.environ\['NUMBAPRO_NVVM'\] = '/usr/local/cuda/nvvm/lib64/libnvvm.so'
os.environ\['NUMBAPRO_LIBDEVICE'\] = '/usr/local/cuda/nvvm/libdevice/'
os.environ\['CONDA_PREFIX'\] = '/usr/local'
!pip uninstall cupy -y

I am getting the following warning:
WARNING conda.core.solve:\_add_specs(639): pinned spec python=3.10 conflicts with explicit specs. Overriding pinned spec.
WARNING conda.core.solve:\_add_specs(639): pinned spec cudatoolkit=11.8 conflicts with explicit specs. Overriding pinned spec.

SpecsConfigurationConflictError: Requested specs conflict with configured specs.
requested specs:
- cudatoolkit=11.2
- dask-sql
- gcsfs
- llvmlite
- mamba==1.4.1
- openssl
- python=3.9
- rapids=22.12
pinned specs:
- python_abi=3.10[build=*cp310*]
Use 'conda config --show-sources' to look for 'pinned_specs' and 'track_features'
configuration parameters. Pinned specs may also be defined in the file
/usr/local/conda-meta/pinned.

and when I am proceeding with the next step:
import cudf

I receive an error:
No module named 'cudf'

Can someone help me with that?

Nata107
  • 31
  • 5
  • on top on nick's suggestion, i've updated RAPIDS version to current latest as this colab env supports latest RAPIDS via conda install: https://github.com/rapidsai-community/rapidsai-csp-utils/pull/79 – TaureanDyerNV May 12 '23 at 23:16

1 Answers1

5

Currently, the preferred way to install RAPIDS frameworks on Colab is to use pip. If you use the "Launch Google Colab (Pip)" link at rapids.ai/#quick-start (or directly run the code in the notebook generated from that link, which I've listed below), things should work for you.

# This get the RAPIDS-Colab install files and test check your GPU.  Run this and the next cell only.
# Please read the output of this cell.  If your Colab Instance is not RAPIDS compatible, it will warn you and give you remediation steps.
!git clone https://github.com/rapidsai/rapidsai-csp-utils.git
!python rapidsai-csp-utils/colab/pip-install.py
Nick Becker
  • 4,059
  • 13
  • 19