AndMastro/EdgeSHAPer
EdgeSHAPer: Bond-Centric Shapley Value-Based Explanation Method for Graph Neural Networks
This tool helps medicinal chemists and materials scientists understand why a Graph Neural Network (GNN) predicts certain properties for a molecule or material. You provide your GNN model and molecular graph data, and it outputs explanations highlighting which chemical bonds (edges) are most important for the prediction. This allows researchers to pinpoint critical structural features contributing to a compound's activity or characteristics.
No commits in the last 6 months.
Use this if you need to interpret the predictions of your Graph Neural Network for molecular or material properties and want to understand the contribution of individual chemical bonds.
Not ideal if your GNN task involves node classification rather than graph classification, or if you are not working with graph-structured data like molecules.
Stars
27
Forks
1
Language
Jupyter Notebook
License
GPL-3.0
Category
Last pushed
Aug 06, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/AndMastro/EdgeSHAPer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
shap/shap
A game theoretic approach to explain the output of any machine learning model.
mmschlk/shapiq
Shapley Interactions and Shapley Values for Machine Learning
iancovert/sage
For calculating global feature importance using Shapley values.
predict-idlab/powershap
A power-full Shapley feature selection method.
aerdem4/lofo-importance
Leave One Feature Out Importance