Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix zero3 save problem with minimum change #1457

Merged
merged 1 commit into from
Jun 5, 2023
Merged

fix zero3 save problem with minimum change #1457

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 15 additions & 3 deletions fastchat/train/train_lora.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -145,10 +145,22 @@ def train():
trainer.train()
trainer.save_state()

# Save states. Weights might be a placeholder in zero3 and need a gather
state_dict = get_peft_state_maybe_zero_3(
# check if zero3 mode enabled
if trainer.hf_deepspeed_config_orig.is_zero3():
# use deepspeed engine internal function to gather state dict
# state_dict_zero3 contains whole parameters of base and lora adapters
# we will not extract lora parameters since peft save_pretrained will do that
# https://github.com/huggingface/peft/blob/3714aa2fff158fdfa637b2b65952580801d890b2/src/peft/peft_model.py#L125
# https://github.com/huggingface/peft/blob/3714aa2fff158fdfa637b2b65952580801d890b2/src/peft/utils/save_and_load.py#L19
state_dict_zero3 = trainer.model_wrapped._zero3_consolidated_16bit_state_dict()
ZYHowell marked this conversation as resolved.
Show resolved Hide resolved
if training_args.local_rank == 0:
state_dict = state_dict_zero3
else:
# in other mode we use original code from fastchat team, to make sure our change is minimum
state_dict = get_peft_state_maybe_zero_3(
model.named_parameters(), lora_args.lora_bias
)
)

if training_args.local_rank == 0:
model.save_pretrained(training_args.output_dir, state_dict=state_dict)

Expand Down