Skip to content

Fix multi-host ImageNet sharding when subset sizes are used #922

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
copybara-service merged 1 commit into from
Jun 30, 2026
Merged

Fix multi-host ImageNet sharding when subset sizes are used #922

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
7 changes: 6 additions & 1 deletion init2winit/dataset_lib/imagenet_dataset.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -309,7 +309,12 @@ def load_split(
# entirely to the end of it on the last host, because otherwise we will drop
# the last `{train,valid}_size % split_size` elements.
if jax.process_index() == jax.process_count() - 1:
end = ''
if split in ['train', 'eval_train']:
end = hps.train_size
elif split == 'valid':
end = hps.valid_size
else:
end = hps.test_size

logging.info('Loaded data [%d: %s] from %s', start, str(end), split)
if split in ['train', 'eval_train']:
Expand Down