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 module initialization for root module under Zero3 #33632

Merged
merged 4 commits into from
Oct 3, 2024
Merged

Fix module initialization for root module under Zero3 #33632

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
059fdb2
Use all state dict keys when checking if root module is initialized.
Ben-Schneider-code Sep 20, 2024
439ca4c
Apply style corrections
Ben-Schneider-code Sep 20, 2024
948aef2
Add comment explaining change.
Ben-Schneider-code Sep 28, 2024
8aa9264
Change comment phrasing.
Ben-Schneider-code Sep 28, 2024
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
3 changes: 3 additions & 0 deletions src/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -608,6 +608,9 @@ def set_initialized_submodules(model, state_dict_keys):
not_initialized_submodules = {}
for module_name, module in model.named_modules():
loaded_keys = {k.replace(f"{module_name}.", "") for k in state_dict_keys if k.startswith(f"{module_name}.")}
# When checking if the root module is loaded all state_dict_keys must be used.
if module_name == "":
loaded_keys = set(state_dict_keys)
if loaded_keys.issuperset(module.state_dict()):
module._is_hf_initialized = True
else:
Expand Down
Loading