search

benob / recasepunc

Model for recasing and repunctuating ASR transcripts
BSD 3-Clause "New" or "Revised" License
126 stars 20 forks source link

missing train.x train.y valid.x valid.y #21

Closed Lorenzoncina closed 10 months ago

Lorenzoncina commented 10 months ago

After having runned stage 2: python recasepunc.py tensorize input.case+punc input.case+punc.x input.case+punc.y --lang $LANG

I can't run stage 3 cause I didn't get train.x train.y valid.x valid.y python recasepunc.py train train.x train.y valid.x valid.y checkpoint/path --lang $LANG

How can I get this files?

benob commented 10 months ago

You have to create a train/valid/test Split and run the command on each of the subset.