Skip to content

fix ddp+graidient_checkpointing bug #61

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

Merged
merged 1 commit into from
Sep 8, 2023
Merged

fix ddp+graidient_checkpointing bug #61

Show file tree
Hide file tree
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
2 changes: 1 addition & 1 deletion examples/pytorch/llm/scripts/seqgpt_560m/full/sft.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ python src/llm_sft.py \
--warmup_ratio 0.03 \
--eval_steps 100 \
--save_steps 100 \
--only_save_model true \
--only_save_model false \
--save_total_limit 2 \
--logging_steps 10 \
--push_to_hub false \
Expand Down
9 changes: 8 additions & 1 deletion examples/pytorch/llm/src/llm_sft.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -296,10 +296,17 @@ def llm_sft(args: SftArguments) -> None:
local_rank=local_rank,
**kwargs)

# trainer_args.ddp_find_unused_parameters = False
if args.gradient_checkpointing:
model.config.use_cache = False
model.enable_input_require_grads()
if is_dist():
if args.gradient_checkpointing:
trainer_args.ddp_find_unused_parameters = False
trainer_args.ddp_broadcast_buffers = False
else:
trainer_args.ddp_find_unused_parameters = True
trainer_args.ddp_broadcast_buffers = True

logger.info(f'trainer_args: {trainer_args}')

trainer = Seq2SeqTrainer(
Expand Down