tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Merge pull request #23 from NanoCode012/patch-1

Browse Source 
Fix: Save adapter for lora
This commit is contained in:
Wing Lian
2023-05-09 15:05:58 -04:00
committed by GitHub
parent df9c5085b5 915c56cd97
commit 1fb0376150
1 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
scripts/finetune.py
View File

@@ -228,8 +228,11 @@ def train(
logging.info( logging.info(
f"Training Completed!!! Saving pre-trained model to {cfg.output_dir}" f"Training Completed!!! Saving pre-trained model to {cfg.output_dir}"
) )
# TODO do we need this fix? https://huggingface.co/docs/accelerate/usage_guides/fsdp#saving-and-loading # TODO do we need this fix? https://huggingface.co/docs/accelerate/usage_guides/fsdp#saving-and-loading
trainer.save_model(cfg.output_dir) trainer.save_pretrained(cfg.output_dir)
# trainer.save_model(cfg.output_dir) # TODO this may be needed for deepspeed to work? need to review another time
if __name__ == "__main__": if __name__ == "__main__":