Thanks to visit codestin.com
Credit goes to github.com

Skip to content

[vision] Add option to process images individually #1706

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

Jump to bottom
Merged
merged 3 commits into from
Jun 2, 2025
Merged

[vision] Add option to process images individually #1706

merged 3 commits into from
Jun 2, 2025

Conversation

@oelachqar
Copy link
Contributor

Description

  • Add option to process images individually:
    • Some processors (e.g. Molmo) do not support processing a batch of images
    • This PR adds an option to process each image individually instead
  • Also add more docstrings / explanations of each option

Related issues

Fixes # (issue)

Before submitting

  • This PR only changes documentation. (You can ignore the following checks in that case)
  • Did you read the contributor guideline Pull Request guidelines?
  • Did you link the issue(s) related to this PR in the section above?
  • Did you add / update tests where needed?

Reviewers

At least one review from a member of oumi-ai/oumi-staff is required.

taenin
taenin reviewed Jun 2, 2025
View reviewed changes
taenin
taenin approved these changes Jun 2, 2025
View reviewed changes
Copy link
Collaborator

@taenin taenin left a comment

Choose a reason for hiding this comment

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

Would love to see tests added, but LGTM as they don't exist yet.

nikg4
nikg4 reviewed Jun 2, 2025
View reviewed changes
src/oumi/core/collators/vision_language_sft_collator.py
max_var_dims = 2 if self._allow_multi_image_inputs else 1

# Pad and stack tensors
collated[key] = pad_to_max_dim_and_stack(
Copy link
Contributor

@nikg4 nikg4 Jun 2, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

note that this simple padding logic isn't compatible with all VLM models. Consider adding a note about it

(delegating everything to processor was the original motivation for adding this collator)

@oelachqar oelachqar merged commit 828466a into main Jun 2, 2025
5 checks passed
@oelachqar oelachqar deleted the oelachqar/process_individually branch June 2, 2025 23:37
penfever pushed a commit that referenced this pull request Aug 27, 2025
@oelachqar
[vision] Add option to process images individually (#1706)
dea3c8c
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@taenin taenin taenin approved these changes

@wizeng23 wizeng23 wizeng23 approved these changes

@jgreer013 jgreer013 Awaiting requested review from jgreer013

+1 more reviewer

@nikg4 nikg4 nikg4 approved these changes

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@oelachqar @taenin @wizeng23 @nikg4