Remove normalization of t euphonique

hipster-philology / nlp-pie-taggers

Extension for pie to include taggers with their models and pre/postprocessors

Mozilla Public License 2.0

11 stars 2 forks source link

Open PonteIneptique opened 4 years ago

PonteIneptique commented 4 years ago

Because models were not trained with -t- on the pronoun but on the verb, we normalize the input

[ ] Lines such as iterator.tokenizer.replacer = lambda x: x # Until model is fixed in test should be removed.
[ ] replacer should be clean`
[ ] Model should be updated