reads y_pred and y_true columns from file entities_f1.csv, and then prints output of classification_report function from sklearn.metrics library
you can use https://github.com/zakcali/class-extractor library for creating y_pred column, from marked html files