“SHAP (SHapley Additive exPlanations) is a unified approach to explain the output of any machine learning model. SHAP connects game theory with local explanations, uniting several previous methods [1-7] and representing the only possible consistent and locally accurate additive feature attribution method based on expectations (see our papers for details and citations).”
https://github.com/slundberg/shap
[protecting people by good design, solid security, efficient processes and trusted services]