Skip to content

support num_workers in EWC #1310

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
WenjinW wants to merge 3 commits into from
+7 −2
Closed

support num_workers in EWC #1310

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5740f23
support num_workers in EWC
WenjinW Mar 3, 2023
050ccd8
Merge branch 'master' into master
WenjinW Mar 3, 2023
620eeec
Update ewc.py
WenjinW Mar 3, 2023
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
9 changes: 7 additions & 2 deletions avalanche/training/plugins/ewc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -121,6 +121,7 @@ def after_training_exp(self, strategy, **kwargs):
strategy.experience.dataset,
strategy.device,
strategy.train_mb_size,
kwargs["num_workers"],
)
self.update_importances(importances, exp_counter)
self.saved_params[exp_counter] = copy_params_dict(strategy.model)
Expand All @@ -129,7 +130,8 @@ def after_training_exp(self, strategy, **kwargs):
del self.saved_params[exp_counter - 1]

def compute_importances(
self, model, criterion, optimizer, dataset, device, batch_size
self, model, criterion, optimizer, dataset, device, batch_size,
num_workers
):
"""
Compute EWC importance matrix for each parameter
Expand All @@ -152,12 +154,15 @@ def compute_importances(

# list of list
importances = zerolike_params_dict(model)

collate_fn = (
dataset.collate_fn if hasattr(dataset, "collate_fn") else None
)
dataloader = DataLoader(
dataset, batch_size=batch_size, collate_fn=collate_fn
dataset, batch_size=batch_size, collate_fn=collate_fn,
num_workers=num_workers
)

for i, batch in enumerate(dataloader):
# get only input, target and task_id from the batch
x, y, task_labels = batch[0], batch[1], batch[-1]
Expand Down