Skip to content

Latest commit

 

History

History
19 lines (12 loc) · 782 Bytes

README.md

File metadata and controls

19 lines (12 loc) · 782 Bytes
Raw

MetaExplainer

Setup

  1. Download DRKG or use the following command:
wget https://dgl-data.s3-us-west-2.amazonaws.com/dataset/DRKG/drkg.tar.gz

  1. When you untar drkg.tar.gz, you will find drkg.tsv.

  2. Add drkg.tsv to Input/DRKG.

Quick Tour

  • TrainGNN.ipynb : Takes as input drkg.tsv and outputs a trained Graph Neural Network (can be either GCN, GraphSAGE or GAT) to Output/GNNModels.
  • ExplainabilityUseCase.ipynb : Takes as input a model and its precitions, and returns a dictionary (for a specific use case) containing the most important nodes that were most influential for the prediction(s) using Saliency Maps.
  • PlotSubgraph.ipynb : Plots the explanatory subgraph.