You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For CI/CD pipeline on Github, I don't want to install GPU dependencies for torch since they're not going to be used anyway, and they make the .venv folder large, thus, make the cache big. My current workaround is to use poetry export to export to requirements.txt & manually removing those dependencies. This is problematic since poetry export is buggy with different versions constraints for the same package (in our case, it is numpy, see #22).
Experimenting with PDM to see if it can solve this problem, and the transition might be easy since it has an option to import from Poetry format.
The text was updated successfully, but these errors were encountered:
For CI/CD pipeline on Github, I don't want to install GPU dependencies for torch since they're not going to be used anyway, and they make the
.venv
folder large, thus, make the cache big. My current workaround is to usepoetry export
to export torequirements.txt
& manually removing those dependencies. This is problematic sincepoetry export
is buggy with different versions constraints for the same package (in our case, it isnumpy
, see #22).Experimenting with PDM to see if it can solve this problem, and the transition might be easy since it has an option to import from Poetry format.
The text was updated successfully, but these errors were encountered: