Scripts for feature extraction, training, and model application.
- param_feature_selection:
- Hyperparameter selection via gridsearch for the 3 classifiers compared in the manuscript.
- Feature selection using Mean Decrease in Accuracy.
- training_evaluation:
- Script to perform cross-validation on resulting models.
- model
- CLI to make predictions using either XGBoost or Random Forest.
- Pandas
- NumPy
- Scikit-learn
- XGBoost
- Keras==2.3.1
- Tensorflow==2.0.0
Zenodo data repository for this work: https://zenodo.org/record/4425690
Link to paper: https://www.pnas.org/content/119/7/e2100916119