Shapash is a Python library released by MAIF data team in January 2021 to make Machine Learning models understandable by everyone. Shapash is currently using a Shap backend to compute local contributions. You will find the general presentation of Shapash in this article. Version 1.3.2 is now available and Shapash allows the Data Scientist to document each model he releases into production. Within a few lines of code, he can include in an HTML report all the information about his model (and its associated performance), the data he uses, his learning strategy, … this report is designed to be easily shared with a Data Protection Officer, an internal audit department, a risk control department, a compliance department or anyone who wants to understand his work.
May-4-2021, 13:40:35 GMT