MLP/Catboost/XGBoost

huangjc

Usage

Put train_labels.csv and train.csv in folder input/predict-student-performance-from-game-play. About 5GB.
Run python preprocess.py
Run the training script train_mlp.py, train_catboost.py and train_xgboost.py
After training, you will see models saved in folder models/
When you have all the models in models, run evaluate.py to make parameter iteration and compute F1 scores (on the full training set)
submit.py is a reference for submission to Kaggle. Note this is a local script, please upload a dataset and correct the model paths for a real online submission.