search

SteveKGYang / MentalLLaMA

This repository introduces MentaLLaMA, the first open-source instruction following large language model for interpretable mental health analysis.
MIT License
192 stars 18 forks source link

MentalBERT #3

Closed vinlincc closed 9 months ago

vinlincc commented 9 months ago

I wonder when fine-tune the MentalBERT, are testing/validation sets feed into the fine-tuning process?

TianlinZhang668 commented 9 months ago

In the downstream task fine-tuning, we do not use test sets.

vinlincc commented 9 months ago

Thank you!