Repository navigation
fix(medcat): Training reset during fine tune - #629
Merged
Merged
Conversation
adam-sutton-1992
approved these changes
Oct 9, 2026
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
During fine tuning (
cat.trainer.train_supervised_raw) the user can opt to reset concept train counts. This is so that the fine-tuning (i.e stuff that's generally been verified by a clinician) has more weight over the vast numbers of unsupervised training the model has seen.However, so far this flow has been broken for datasets with concepts that are not in the model / CDB.
This PR fixes that by only resetting cui count for concepts in the CDB.
It also adds a few tests, some of which will fail upon the first commit the workflow runs on to demonstrate the issue.
The subsequent commit will provide the actual patch to fix the issue.