The model uses a pre-trained model which classifies the text into the following categories: scientific, journalistic, beletristic, memorialistic and judicial.
You can download the pre-trained model from here: http://speech.utcluj.ro/data/cnn_text_classif_5classes_new_punct.h5
Questions? adriana.stan@com.utcluj.ro