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

fix bug of i2v finetune #657

Merged
merged 3 commits into from
Jan 12, 2025
Merged

fix bug of i2v finetune #657

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
35383e2
fix potential bug of i2v
ZGCTroy Jan 11, 2025
cd861bb
Update i2v_dataset.py
ZGCTroy Jan 11, 2025
09a49d3
fix bug of i2v; video is already 0-255
ZGCTroy Jan 11, 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
1 change: 1 addition & 0 deletions finetune/datasets/i2v_dataset.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,6 +148,7 @@ def __getitem__(self, index: int) -> Dict[str, Any]:
frames, image = self.preprocess(video, image)
frames = frames.to(self.device)
image = image.to(self.device)
image = self.image_transform(image)
Copy link
Collaborator

Choose a reason for hiding this comment

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

Please also transform the image after line 146, thank you ~

Copy link
Collaborator

Choose a reason for hiding this comment

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

Alright, to speed up the process, I'll add this line of code myself. Thank you for your contribution.

# Current shape of frames: [F, C, H, W]
frames = self.video_transform(frames)

Expand Down
2 changes: 1 addition & 1 deletion finetune/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -526,7 +526,7 @@ def validate(self, step: int) -> None:
video, self.state.train_frames, self.state.train_height, self.state.train_width
)
# Convert video tensor (F, C, H, W) to list of PIL images
video = (video * 255).round().clamp(0, 255).to(torch.uint8)
video = video.round().clamp(0, 255).to(torch.uint8)
video = [Image.fromarray(frame.permute(1, 2, 0).cpu().numpy()) for frame in video]

logger.debug(
Expand Down