Machine Learning Pipeline

The whole pipeline is prototyped in prototype.ipynb.

The updated version is split into multiple files. data_preparation.ipynb contains code for data preparation from raw transactional data to a prepared csv file containing X and y for training. training.ipynb runs the actual training process with different configurations. feature_evaluation.ipynb contains evaluation for the calculated metrics features.