Developing a Fidelity Evaluation Approach for Interpretable Machine Learning

This repository contains code associated with the article "Developing a Fidelity Evaluation Approach for Interpretable Machine Learning" by Mythreyi Velmurugan, Chun Ouyang, Catarina Moreira and Renuka Sindhgatta. The code in this repository can be used to evaluate the explanations of LIME and SHAP for decision tree and XGBoost models trained on tabular data.

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
.ipynb_checkpoints		.ipynb_checkpoints
bike_sharing		bike_sharing
breast_cancer		breast_cancer
diabetes		diabetes
housing		housing
income		income
student_scores		student_scores
README.md		README.md
check_quartile_decile_distributions.ipynb		check_quartile_decile_distributions.ipynb
compile_results.ipynb		compile_results.ipynb
diabetes_tree.eps		diabetes_tree.eps
find_deciles_and_scale.ipynb		find_deciles_and_scale.ipynb
generate_datasets.ipynb		generate_datasets.ipynb
generate_design_stats.ipynb		generate_design_stats.ipynb
generate_model.ipynb		generate_model.ipynb
test_fidelity.ipynb		test_fidelity.ipynb
test_white_box_fidelity.ipynb		test_white_box_fidelity.ipynb

Mythreyi-V/three-phase-fidelity-evaluation

Folders and files

Latest commit

History

Repository files navigation

Developing a Fidelity Evaluation Approach for Interpretable Machine Learning

About

Resources

Stars

Watchers

Forks

Languages