0
$\begingroup$

Below is an example of classification report output by sklearn. Is there an easy way to identify which features contribute to the precision and recall improvement for each label? I can start with using a single feature to build a model and add features one by one to see how the precision and recall of each label improve in the classification report. But obviously, this approach is quite tedious and time consuming plus may not work.

              precision    recall  f1-score   support

          C0       0.38      0.65      0.48      4484
          C1       0.23      0.34      0.27      3617
          C2       0.26      0.38      0.31      3221
          C3       0.23      0.17      0.19      2535
          C4       0.50      0.45      0.47      2370
          C5       0.31      0.23      0.26      2161

    accuracy                           0.32     26735
   macro avg       0.35      0.22      0.24     26735
weighted avg       0.32      0.32      0.30     26735
```
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.