icdm_graph_competition/pyg_example

• PyG example
• • • Include
    • How to run

PyG example

Include

• pyg_demo.sh: Shell Script (include converting raw data into PyG graph, training model and generating final result)
• rgcn_mb_icdm.py: PyG rgcn model
• format_pyg.py: PyG data generator
• best_model/: Directory to store model trained

How to run

• Run shell script

mkdir -p /dataset/pyg_data/
cd /code/icdm_graph_competition/pyg_example/
sh pyg_demo.sh

• Output

  1. PyG Graph is stored in /dataset/pyg_data/

     • icdm2022_session1.pt PyG Heterograph of session1 data

     • icdm2022_session2.pt PyG Heterograph of session2 data

     • *.nodes.pyg Temporary cache file (Removable)

  2. Trained Model is saved in /code/icdm_graph_competition/best_model/

     • (Default) 1.pth Model file
     • (Optional) change model_id to customize name of model file

  3. Final result of Inference is generated in /code/icdm_graph_competition/pyg_example/

     • pyg_session1.json Result of session1
     • pyg_session2.json Result of session2