Which T5 model was used for fine-tuning?

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

Which T5 model was used for fine-tuning? #11

Closed Joapfel closed 2 years ago

Joapfel commented 2 years ago

Hi, could you kindly tell which T5 model (T5-small, T5-base, T5-large, T5-3B, ....) the Parrot Project is based on?

Kind regards

TiloMichel commented 2 years ago

It seems to be the T5-base model because the model from Parrot has the same size (pytorch-model.bin 850 MB) as the T5-base model.

Sources: Parrot Model: https://huggingface.co/prithivida/parrot_paraphraser_on_T5/tree/main T5-base: https://huggingface.co/t5-base/tree/main