Which dataset did you use to train?
#1
by
ashutoshsaboo
- opened
Interesting that you used mordernbert for punctuation restoration task. I've been in the process of training similar models (which I will open source soon) and wanted to baseline against yours. Which dataset did you pretrain your model on? @whooray
hi,
@ashutoshsaboo
i trained the model using
Training data: libriheavy & mls-eng datasets
Evaluation data: mls-eng dataset
thanks for asking.