hosseinsarshar
commented
1 year ago - Working on Captum - there were some issues with multi-class models - it currently only reports on three classes - Positive, Neutral, and Negative.
- Trying the text explainability library of interpret.ml at: https://github.com/interpretml/interpret-text