TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration

Open in new window