hotfix(medcat): Disable config train at load time#553
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 PR sets
config.components.linking.traintoFalseat load time.Some models may have been saved with this set to
True. And as of v2.8 (specifically #414) this can cause issues since that method of running training is no longer used.There's also an added test that makes sure the loaded model has this config option disabled. (NOTE: the model that's in the repo has this problem so the test previously set this to
Falseitself, but I've now removed this from the test class setup).I also checked that this test fails in the main branch (i.e without the change from the first commit).
PS:
Just to be clear, this has to do with the config option that was previously used for self-supervised training. The process was to set
*.train = Trueand run inference and then the vocab based linker would perform training within the inference step. The expected new approach is more explicit and does not use this config option. So all models can still be trained - both unsupervised and supervised.