Skip to content

[checkpointio] gather tensor before unpad it if the tensor is both padded and distributed #6168

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
Jan 21, 2025
Merged

[checkpointio] gather tensor before unpad it if the tensor is both padded and distributed #6168

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
4 changes: 2 additions & 2 deletions colossalai/checkpoint_io/hybrid_parallel_checkpoint_io.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,9 +107,9 @@ def _model_sharder(
if param is None:
continue
# Gather tensor pieces when using tensor parallel.
if is_padded_tensor(param):
param = to_unpadded_tensor(param)
param_ = gather_distributed_param(param, keep_vars=False)
if is_padded_tensor(param_):
param_ = to_unpadded_tensor(param_)
if pinned_state_dicts is not None:
if (prefix + name) not in pinned_state_dicts:
pinned_state_dicts[prefix + name] = torch.empty_like(param_, pin_memory=True, device="cpu")
Expand Down
Loading