In the token loop, token.text is the original word string, token.lemma_ is the base form of the word, token.pos_ is the simple part-of-speech tag, token.dep_ is the syntactic dependency relation, and token.is_stop is a boolean indicating if the token is a stop word or not.

In the entity loop, ent.text is the original entity string, ent.start_char and ent.end_char are the start and end indices of the entity in the original string, and ent.label_ is the type of the entity.

Please make sure to download the language models first, using python -m spacy download en_core_web_sm, for example.

Note that Spacy’s capabilities go well beyond this, including text classification, similarity detection, rule-based matching, and integration with machine learning libraries like TensorFlow and PyTorch.