CLG-lab-CS-Clf

code-switch classifier - next token predication

First time, do the following:

1. extract bm_tagged/bm_tagged.zip
2. create bm_tagged/ (or bm_tagged_w_cognatehood/) and run yuli_cognates/tag_cognatehood.py to get the .csv files
3. create corpus/ and run read_corpus.py to get the .dat files
4. run & explore the uter_cls.ipynb