Data is only available on kaggle currently. In order to run example code, you need to download data from kaggle and put into data folder in project root. See examples/pairing.py for more detail.
Please add project root to your $PYTHONPATH or virtual environment package path in order to import package properly. If you are using PyCharm, this should be default behavior.
- SciKit-learn
- Pandas
- PyTorch
- Matplotlib
Data preprocessing is implement in cpt_data_preprocessing. See examples/filters.py and examples/pairing.py for more detail.
GNN model is implement in cpt_gnn. Training option can be found in examples/configs. See examples/gnn.py for more detail.
Note that you should run examples/pairing.py first to generate pairing data.
Visualization is implement in examples/cpt_plots. It provide many drawing method for different situation. See examples/gnn.py and examples/plot_500_particles.py for more detail.