ModuleNotFoundError: No module named 'pkg_resources' when installing conda environment with make install

[SamBryce-Smith]

The provided conda environment does not ensure setuptools is installed, causing a module not found error. Adding setuptools<82.0.0 to environment.yml resolves the error on my end:

$ make install
...

Transaction finished


To activate this environment, use:

    mamba activate RiboNN

Or to execute a single command in this environment, use:

    mamba run -n RiboNN mycommand

$ make predict_human 
python3 -m src.main --predict human
Traceback (most recent call last):
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "/lustre1/project/stg_00130/Projects/sbs_config/github_repos/Sanofi-Public/RiboNN/src/main.py", line 5, in <module>
    from src.train import train_model_nested_cv
  File "/lustre1/project/stg_00130/Projects/sbs_config/github_repos/Sanofi-Public/RiboNN/src/train.py", line 4, in <module>
    import pytorch_lightning as pl
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/site-packages/pytorch_lightning/__init__.py", line 34, in <module>
    from lightning_lite.utilities.seed import seed_everything  # noqa: E402
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/site-packages/lightning_lite/__init__.py", line 23, in <module>
    from lightning_lite.lite import LightningLite  # noqa: E402
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/site-packages/lightning_lite/lite.py", line 23, in <module>
    from lightning_utilities.core.apply_func import apply_to_collection
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/site-packages/lightning_utilities/__init__.py", line 6, in <module>
    from lightning_utilities.core.apply_func import apply_to_collection
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/site-packages/lightning_utilities/core/__init__.py", line 4, in <module>
    from lightning_utilities.core.imports import compare_version, module_available
  File "/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/site-packages/lightning_utilities/core/imports.py", line 14, in <module>
    import pkg_resources
ModuleNotFoundError: No module named 'pkg_resources'
make: *** [Makefile:21: predict_human] Error 1

After removing the installed environment, adding setuptools<82.0.0 to the environment.yml and rerunning make install, the error is resolved and I get a successful test run on my HPC:

$ make install
...
Transaction finished


To activate this environment, use:

    mamba activate RiboNN

Or to execute a single command in this environment, use:

    mamba run -n RiboNN mycommand
$ mamba activate RiboNN
$ make predict_human 
python3 -m src.main --predict human
/lustre1/project/stg_00130/Projects/sbs_config/miniforge/vsc37853/envs/RiboNN/lib/python3.10/site-packages/lightning_utilities/core/imports.py:14: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
  import pkg_resources
Making predictions using human models...
Transcripts with combined CDS and 3'UTR sizes more than 11937 in the input data are removed!
Predictions written to results/human/prediction_output.txt.

[SamBryce-Smith]

I've also made a conda-lock conda environment for anyone who wants to exactly reproduce a conda environment (linux-64 platform only). Have successfully tested with make predict_human

conda-lock.yml

conda-lock install --name RiboNN conda-lock.yml or micromamba create -n RiboNN -f conda-lock.yml