Returns the document frequencies, i.e. the number of documents that contain term.
the array of document frequencies.
the token list used as features.
the training corpus.