Prediction on new data set

To make predictions on new dataset with the full model, run:

proms_predict -m /path/to/saved/full_model.pkl -d /path/to/data_file

The data file should have the same format as Training data. The file should contain measurments for all markers selected by the full model.

An output file (default name: prediction_output.txt) will be generated after the prediction. For classification task, it contains 3 columns: sample, probability (of being class 1), (predicted) label.