Notes on Applicability of Explainable AI Methods to Machine Learning Models Using Features Extracted by Persistent Homology