search

marrlab / DomainLab

modular domain generalization: https://pypi.org/project/domainlab/
https://marrlab.github.io/DomainLab/
MIT License
42 stars 2 forks source link

fbopt ci failed after merged master into fbopt, with pr commandline interface entry point #754

Closed smilesun closed 8 months ago

smilesun commented 8 months ago

https://github.com/marrlab/DomainLab/actions/runs/7583966552/job/20656686984#step:7:786 with commit https://github.com/marrlab/DomainLab/pull/462/commits/aa1d5cd13e6a9db9d829604f81cdc1d7570876cb

no module named tensorboard

the previous commit has no such a problem.

can it be the pyproject.yaml file? https://github.com/marrlab/DomainLab/pull/740/files

smilesun commented 8 months ago

@agisga , will we get inconsistency now for the requirements.txt which is required by pypi and the pyproject.toml??

https://github.com/marrlab/DomainLab/pull/462/commits/2319105e39644757beba7d8aade32ed73576ec8c

smilesun commented 8 months ago

now Creating virtualenv domainlab-okngg5wM-py3.10 in /home/runner/.cache/pypoetry/virtualenvs Installing dependencies from lock file Warning: poetry.lock is not consistent with pyproject.toml. You may be getting improper dependencies. Run poetry lock [--no-update] to fix it.

Because domainlab depends on tensorboard (^2.14.0) which doesn't match any versions, version solving failed. Error: Process completed with exit code 1.

https://github.com/marrlab/DomainLab/actions/runs/7584309962/job/20657710847#step:6:132

smilesun commented 8 months ago

this commit revert the way how pytest is run

https://github.com/marrlab/DomainLab/pull/462/commits/04a6f25da58c489a1c66cccb6ad2c8885495a4ec

agisga commented 8 months ago

@agisga , will we get inconsistency now for the requirements.txt which is required by pypi and the pyproject.toml??

2319105

requirements.txt was generated from pyproject.toml before. Did you rerun https://github.com/marrlab/DomainLab/blob/master/scripts/sh_gen_requirements.sh ?

agisga commented 8 months ago

tensorboard

It's better to run poetry add tensorboard instead of editing pyproject.toml by hand. Then it should find a compatible version. You may also need to run poetry lock --no-update if you still get the warning about poetry.lock.

smilesun commented 8 months ago

@agisga , do you have time to take a look maybe? I tried several commits but still could not resovle it.

agisga commented 8 months ago

@agisga , do you have time to take a look maybe? I tried several commits but still could not resovle it.

I can take a look at it tomorrow.

smilesun commented 8 months ago

thanks! that will be great!

agisga commented 8 months ago

thanks! that will be great!

The PR seems to address the issue, unless I'm misunderstanding something: https://github.com/marrlab/DomainLab/pull/765

smilesun commented 8 months ago

thanks! that will be great!

The PR seems to address the issue, unless I'm misunderstanding something: #765

thanks! How did you find out the version fo tensorboard should be the one you changed to in the pyproject.toml?

https://github.com/marrlab/DomainLab/pull/462/commits/486d81a1e0c269725d6058faf81e31c474e1fa96

agisga commented 8 months ago

thanks! How did you find out the version fo tensorboard should be the one you changed to in the pyproject.toml?

I first deleted tensorboard from pyproject.toml, and then ran poetry add tensorboard. Then poetry determined this specific version to be used. I'm not sure how Poetry decides on what version to use, but I assume that it checks for compatibility with the other packages specified.