Here is my code. !pip install cudf

Collecting cudf
  Using cached cudf-0.6.1.post1.tar.gz (1.1 kB)
  Preparing metadata (setup.py) ... done
Building wheels for collected packages: cudf
  error: subprocess-exited-with-error
  
  × python setup.py bdist_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.
  Building wheel for cudf (setup.py) ... error
  ERROR: Failed building wheel for cudf
  Running setup.py clean for cudf
Failed to build cudf
ERROR: Could not build wheels for cudf, which is required to install pyproject.toml-based projects
 `pip install cudf-cuda115`
ERROR: Could not find a version that satisfies the requirement cudf-cuda115 (from versions: none)
ERROR: No matching distribution found for cudf-cuda115
`%load_ext cudf.pandas
import pandas as pd`

`pip install --extra-index-url=https://pypi.nvidia.com cudf-cu12`
Looking in indexes: https://pypi.org/simple, https://pypi.nvidia.com
Collecting cudf-cu12
  Downloading https://pypi.nvidia.com/cudf-cu12/cudf_cu12-23.12.1-cp310-cp310-manylinux_2_28_x86_64.whl (511.6 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 511.6/511.6 MB 2.8 MB/s eta 0:00:00
Requirement already satisfied: cachetools in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (5.3.2)
Collecting cuda-python<13.0a0,>=12.0 (from cudf-cu12)
  Downloading cuda_python-12.3.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (23.6 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 23.6/23.6 MB 27.5 MB/s eta 0:00:00
Requirement already satisfied: cupy-cuda12x>=12.0.0 in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (12.2.0)
Requirement already satisfied: fsspec>=0.6.0 in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (2023.6.0)
Collecting numba<0.58,>=0.57 (from cudf-cu12)
  Downloading numba-0.57.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (3.6 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.6/3.6 MB 38.1 MB/s eta 0:00:00
Requirement already satisfied: numpy<1.25,>=1.21 in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (1.23.5)
Collecting nvtx>=0.2.1 (from cudf-cu12)
  Downloading nvtx-0.2.8-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (582 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 582.4/582.4 kB 31.4 MB/s eta 0:00:00
Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (23.2)
Requirement already satisfied: pandas<1.6.0dev0,>=1.3 in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (1.5.3)
Collecting protobuf<5,>=4.21 (from cudf-cu12)
  Downloading protobuf-4.25.1-cp37-abi3-manylinux2014_x86_64.whl (294 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 294.6/294.6 kB 24.5 MB/s eta 0:00:00
Collecting pyarrow<15.0.0a0,>=14.0.1 (from cudf-cu12)
  Downloading pyarrow-14.0.2-cp310-cp310-manylinux_2_28_x86_64.whl (38.0 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 38.0/38.0 MB 12.3 MB/s eta 0:00:00
Requirement already satisfied: rich in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (13.7.0)
Collecting rmm-cu12==23.12.* (from cudf-cu12)
  Downloading https://pypi.nvidia.com/rmm-cu12/rmm_cu12-23.12.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.0 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.0/2.0 MB 63.7 MB/s eta 0:00:00
Requirement already satisfied: typing-extensions>=4.0.0 in /usr/local/lib/python3.10/dist-packages (from cudf-cu12) (4.5.0)
Requirement already satisfied: fastrlock>=0.5 in /usr/local/lib/python3.10/dist-packages (from cupy-cuda12x>=12.0.0->cudf-cu12) (0.8.2)
Collecting llvmlite<0.41,>=0.40.0dev0 (from numba<0.58,>=0.57->cudf-cu12)
  Downloading llvmlite-0.40.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (42.1 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 42.1/42.1 MB 12.1 MB/s eta 0:00:00
Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas<1.6.0dev0,>=1.3->cudf-cu12) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas<1.6.0dev0,>=1.3->cudf-cu12) (2023.3.post1)
Requirement already satisfied: markdown-it-py>=2.2.0 in /usr/local/lib/python3.10/dist-packages (from rich->cudf-cu12) (3.0.0)
Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /usr/local/lib/python3.10/dist-packages (from rich->cudf-cu12) (2.16.1)
Requirement already satisfied: mdurl~=0.1 in /usr/local/lib/python3.10/dist-packages (from markdown-it-py>=2.2.0->rich->cudf-cu12) (0.1.2)
Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas<1.6.0dev0,>=1.3->cudf-cu12) (1.16.0)
Installing collected packages: nvtx, cuda-python, pyarrow, protobuf, llvmlite, numba, rmm-cu12, cudf-cu12
  Attempting uninstall: pyarrow
    Found existing installation: pyarrow 10.0.1
    Uninstalling pyarrow-10.0.1:
      Successfully uninstalled pyarrow-10.0.1
  Attempting uninstall: protobuf
    Found existing installation: protobuf 3.20.3
    Uninstalling protobuf-3.20.3:
      Successfully uninstalled protobuf-3.20.3
  Attempting uninstall: llvmlite
    Found existing installation: llvmlite 0.41.1
    Uninstalling llvmlite-0.41.1:
      Successfully uninstalled llvmlite-0.41.1
  Attempting uninstall: numba
    Found existing installation: numba 0.58.1
    Uninstalling numba-0.58.1:
      Successfully uninstalled numba-0.58.1
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
ibis-framework 6.2.0 requires pyarrow<13,>=2, but you have pyarrow 14.0.2 which is incompatible.
tensorboard 2.15.1 requires protobuf<4.24,>=3.19.6, but you have protobuf 4.25.1 which is incompatible.
tensorflow-metadata 1.14.0 requires protobuf<4.21,>=3.20.3, but you have protobuf 4.25.1 which is incompatible.
Successfully installed cuda-python-12.3.0 cudf-cu12-23.12.1 llvmlite-0.40.1 numba-0.57.1 nvtx-0.2.8 protobuf-4.25.1 pyarrow-14.0.2 rmm-cu12-23.12.0

I tried more lines of installation but have overwritten them so I don't have them. In short I have tried it all. I am expecting the proper Cudf package with no errors like unknown attribute uint8 in Cupy that comes even after Cudf is installed. Reiterating I want it without conda I am running all latest versions on colab and am willing to downgrade.`

1

There are 1 best solutions below

0
Bradley Dice On

Please refer to the RAPIDS Installation Guide for general installation commands and supported Python/CUDA/OS versions. https://docs.rapids.ai/install

From that, the command to be used for pip is:

pip install --extra-index-url=https://pypi.nvidia.com cudf-cu12==23.12.*

Your install command output shows a conflict between the Arrow versions required by Ibis and the Arrow versions required by cuDF.

For trying cuDF with Colab, try the notebooks linked on the RAPIDS home page: https://rapids.ai/

The Colab notebook linked there should handle any dependency conflicts in its setup cells. These conflicts change frequently due to Colab updates so I don’t think I can describe them in a stable / permanent answer.