Fix weights for regression for non-binned features

thomasp85 / lime

Local Interpretable Model-Agnostic Explanations (R port of original Python package)

https://lime.data-imaginist.com/

Other

481 stars 109 forks source link

Fix weights for regression for non-binned features #110

Closed samleegithub closed 5 years ago

samleegithub commented 6 years ago

Multiply weight of non-binned features by their values. For regressions, change labels to Positive and Negative. Also fix the x-axis scale to make it easier to visually compare across all cases.

samleegithub commented 6 years ago

This implements the suggestion in item 1 of this issue: https://github.com/thomasp85/lime/issues/68

thomasp85 commented 5 years ago

Thanks