search

WayScience / phenotypic_profiling

Machine learning for predicting 15 single-cell phenotypes from cell morphology profiles
Creative Commons Attribution 4.0 International
2 stars 3 forks source link

Add Single Class Model PR Curves #26

Closed roshankern closed 1 year ago

roshankern commented 1 year ago

In this pull request, precision-recall curves are added for the single-class models trained in #25. A PR curve is made for each label type (eg positive and negative label), for each model (Large single-class model, Metaphase single-class model, etc), for each evaluation dataset (train or test), for each feature type (CP, DP, CP_and_DP).

These PR curve figures have been added to class_PR_curves.ipynb and the PR data have been saved to precision_recall_curves.

review-notebook-app[bot] commented 1 year ago

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

roshankern commented 1 year ago

Unfortunately, during #25 (training single class models) all notebooks were run in my base python environment instead of the correct phenotypic_profiling environment. Thus, I reran those notebooks in the correct environment here as well. Please ignore all the recreated model files (ending in .joblib) by scrolling to the bottom of the pull request review.

roshankern commented 1 year ago

This PR is ready for review!