search

emanjavacas / pie

A fully-fledge PyTorch package for Morphological Analysis, tailored to morphologically rich and historical languages.
MIT License
22 stars 10 forks source link

Reworked it a little bit to make preprocessing more homogeneous #60

Closed emanjavacas closed 4 years ago

emanjavacas commented 4 years ago

This should add the "char_lower" and "word_lower" options for lowering input. I also reworked the pipeline so adding stuff to label_encoder.preprocess should work just fine.

emanjavacas commented 4 years ago

All tests run fine, but it'd be nice to run it first through a couple of training/evaluation iterations to make sure old models work as well as newer ones.

emanjavacas commented 4 years ago

Seems to be working fine.