Skip to content

Changed the way the optimizer is saved: only its state_dict is pickled #772

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

Closed
arturshagito0 wants to merge 1 commit into from
Closed

Changed the way the optimizer is saved: only its state_dict is pickled #772

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
6 changes: 4 additions & 2 deletions batchflow/models/torch/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1747,8 +1747,10 @@ def save(self, path, use_onnx=False, path_onnx=None, use_openvino=False, path_op
path, pickle_module=pickle_module, **kwargs)

else:
torch.save({item: getattr(self, item) for item in self.PRESERVE},
path, pickle_module=pickle_module, **kwargs)
attributes = {item: getattr(self, item) for item in self.PRESERVE if item != "optimizer"}
optimizer = getattr(self, "optimizer")
attributes["optimizer"] = optimizer.state_dict()
Copy link
Member

@AlexeyKozhevin AlexeyKozhevin Jan 20, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

But how do we load the optimizer from that state if we need to? In the current implementation we dump Optimizer instance and can deserialize it to use.

torch.save(attributes, path, pickle_module=pickle_module, **kwargs)

def load(self, file, make_infrastructure=False, mode='eval', pickle_module=dill, **kwargs):
""" Load a torch model from a file.
Expand Down
Loading