1. run the train.py code to train the models.
2. The models will be saved as pickle files and the performance of the models will be given as output.
3. Load the best performing model (the pickle files), to test the model on test data.
4. Code for testing is also present in train.py file (lines 89-94).