This repository documents the learnings from completing the Machine Learning Explainability course on Kaggle.
Hands-on project focused on extracting human-understandable insights from any model.
-
lesson_2: permutation importance; what features the model thinks are important. -
lesson_3: partial plots; how each feature affects the preditions. -
lesson_4: SHAP values; understanding individual predictions. -
lesson_5: advanced uses of SHAP values; aggregating SHAP values for even more detailed model insights.
Distributed under the MIT License. See LICENSE.txt for more information.
Leonardo Santos - leorsantos2003@gmail.com