[BUG] Installation Error for transformers4rec

[silpara]

Bug description

Error when installing transformers4rec following instructions from the repo main page.

Steps/Code to reproduce bug

conda create -n transformers4rec-23.04 -c nvidia -c rapidsai -c pytorch -c conda-forge \
    transformers4rec=23.04 `# NVIDIA Merlin` \
    nvtabular=23.04 `# NVIDIA Merlin - Used in example notebooks` \
    python=3.10 `# Compatible Python environment` \
    cudf=23.02 `# RAPIDS cuDF - GPU accelerated DataFrame` \
    cudatoolkit=11.8 pytorch-cuda=11.8 `# NVIDIA CUDA version`

results in error

Collecting package metadata (current_repodata.json): done
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failed

PackagesNotFoundError: The following packages are not available from current channels:

  - nvtabular=23.04
  - cudf=23.02

Current channels:

  - https://conda.anaconda.org/nvidia/win-64
  - https://conda.anaconda.org/nvidia/noarch
  - https://conda.anaconda.org/rapidsai/win-64
  - https://conda.anaconda.org/rapidsai/noarch
  - https://conda.anaconda.org/pytorch/win-64
  - https://conda.anaconda.org/pytorch/noarch
  - https://conda.anaconda.org/numba/win-64
  - https://conda.anaconda.org/numba/noarch
  - https://conda.anaconda.org/conda-forge/win-64
  - https://conda.anaconda.org/conda-forge/noarch
  - https://repo.anaconda.com/pkgs/main/win-64
  - https://repo.anaconda.com/pkgs/main/noarch
  - https://repo.anaconda.com/pkgs/r/win-64
  - https://repo.anaconda.com/pkgs/r/noarch
  - https://repo.anaconda.com/pkgs/msys2/win-64
  - https://repo.anaconda.com/pkgs/msys2/noarch

To search for alternate channels that may provide the conda package you're
looking for, navigate to

    https://anaconda.org

and use the search bar at the top of the page.

Expected behavior

Successful installation.

Environment details

• Platform: Windows 11

[rnyak]

@silpara Hello.. what's your cuda toolkit and driver versions? can you print out nvcc --version and nvidia-smi commands and share with us.

also can you try replacing cudf=23.02 with cudf=22.12 and note that you need dask_cudf as well.

Besides, you dont have to use conda to install Merlin or cudf. Merlin has docker images which helps you to get started right away. In addition, you can install Rapids and Merlin via pip. To help you there we need to know your GPU HW specs, driver version, cuda toolkit version and if you are trying to install Merlin on a Cloud instance or on an on prem WS? thanks.

[rnyak]

@silpara I am closing this ticket since we did not hear back from you. you can reopen if you have any question.