Description
Your model predicts well, but you can't explain why it flagged a customer as high risk. SHAP uses game-theoretic Shapley values to explain the output of any machine learning model.
It computes each feature's contribution to a prediction and plots waterfalls and beeswarms, from tree models to deep learning.
Attribution:Shapley values.
Plots:Waterfall and beeswarm.
Any model:Trees to deep nets.
Views:Global and local.
It computes each feature's contribution to a prediction and plots waterfalls and beeswarms, from tree models to deep learning.
Features
Attribution:Shapley values.
Plots:Waterfall and beeswarm.
Any model:Trees to deep nets.
Views:Global and local.
