fix: improve handling of missing torch dependency in pipeline initial… #6
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This pull request makes improvements to the
TransformersModeratorintegration, specifically in the model loading process to better handle dynamic imports and dependencies. The main change ensures that if PyTorch (torch) is loaded during runtime and wasn't present before, thetransformersmodule is reloaded to avoid potential issues with framework detection.Dependency and dynamic import handling:
src/moderators/integrations/transformers_moderator.py, added logic to check iftorchwas loaded during runtime, and if so, reloads thetransformersmodule to ensure proper framework detection and compatibility.importlibandsysto support dynamic reloading and module inspection for dependency management.