Instructions

Warning: This code may be messy and could contain bugs. This code is just for reference.

1. Download the Data

Download the dataset from Stanford PURL and extract the files. Ensure you have the S1~10.mat files available.

2. Preprocess the Data

Navigate to the data directory and run the following script to generate .npy files:

cd data
python preprocess_v2.py

This will create S1~10.npy.

3. Run Experiments

Navigate to the training directory and execute the experiment scripts:

cd training
python exp1.py
python exp2.py

These scripts reproduce the experiments described in the paper.

4. Explainability Analysis

Navigate to the explainer directory and run the following scripts in order:

1. Generate XAI Input Data

   cd explainer
   python genXaiInput_all.py

2. Run GNN Explainer

   python GNNexplainer_all.py

3. Plot Explainability Results

   python plotXai.py

This process will generate and visualize explainability results.