How was the fluency model trained?

PrithivirajDamodaran / Parrot_Paraphraser

A practical and feature-rich paraphrasing framework to augment human intents in text form to build robust NLU models for conversational engines. Created by Prithiviraj Damodaran. Open to pull requests and other forms of collaboration.

Apache License 2.0

867 stars 143 forks source link

How was the fluency model trained? #3

Closed sajjjadayobi closed 3 years ago

sajjjadayobi commented 3 years ago

As I understand, this is just a BERT with a binary classification head. If that is the case, then what was the training data for this model?

PrithivirajDamodaran commented 3 years ago

This is not an issue, please use this forum for creating issues ONLY