Graph-COM/GSAT
[ICML 2022] Graph Stochastic Attention (GSAT) for interpretable and generalizable graph learning.
When analyzing complex connected data like social networks or molecular structures, it's often hard to understand *why* a machine learning model makes certain predictions. This project helps scientists and data analysts pinpoint the most crucial parts of a graph that drive a model's decision, making its behavior transparent. You input a graph and a model's prediction, and it highlights the key nodes and connections responsible for that outcome, allowing you to trust and generalize your findings.
174 stars. No commits in the last 6 months.
Use this if you need to understand the underlying patterns and critical features within graph-structured data that influence your predictive models, rather than just getting a prediction.
Not ideal if your primary goal is simply to achieve the highest predictive accuracy without needing insight into how the model arrived at its conclusions.
Stars
174
Forks
24
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Feb 19, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Graph-COM/GSAT"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pyg-team/pytorch_geometric
Graph Neural Network Library for PyTorch
a-r-j/graphein
Protein Graph Library
raamana/graynet
Subject-wise networks from structural MRI, both vertex- and voxel-wise features (thickness, GM...
pykale/pykale
Knowledge-Aware machine LEarning (KALE): accessible machine learning from multiple sources for...
dmlc/dgl
Python package built to ease deep learning on graph, on top of existing DL frameworks.