Skip to content

Fix kontext finetune issue when batch size >1 #11921

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 8 commits into from
Jul 18, 2025
Merged

Fix kontext finetune issue when batch size >1 #11921

Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
fccd300
set drop_last to True
mymusise Jul 14, 2025
46ef36d
Merge branch 'main' into main
mymusise Jul 15, 2025
f454cbc
Merge branch 'main' into main
sayakpaul Jul 15, 2025
2d2bef8
Merge branch 'main' into main
asomoza Jul 16, 2025
9c82389
Merge branch 'main' into main
linoytsaban Jul 16, 2025
a37d070
Merge branch 'main' into main
asomoza Jul 18, 2025
9b3e3eb
Merge branch 'main' into main
asomoza Jul 18, 2025
ad65537
Merge branch 'main' into main
asomoza Jul 18, 2025
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
2 changes: 1 addition & 1 deletion examples/dreambooth/train_dreambooth_lora_flux_kontext.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1614,7 +1614,7 @@ def load_model_hook(models, input_dir):
)
if args.cond_image_column is not None:
logger.info("I2I fine-tuning enabled.")
batch_sampler = BucketBatchSampler(train_dataset, batch_size=args.train_batch_size, drop_last=False)
batch_sampler = BucketBatchSampler(train_dataset, batch_size=args.train_batch_size, drop_last=True)
train_dataloader = torch.utils.data.DataLoader(
train_dataset,
batch_sampler=batch_sampler,
Expand Down
Loading