open-compass / open-compass/VLMEvalKit
Evaluating Ola-7b RuntimeError: mat1 and mat2 shapes cannot be multiplied (16x2304 and 1152x3584)
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 4.4k
- Forks
- 768
- Avg merge
- 1d 10h
- Merged PRs (30d)
- 17
Description
The error roots in img_feat = self.get_model().mm_projector(lowres_img_features[idx], lowres_img_sizes[idx], highres_img_features[idx], highres_img_sizes[idx], modalities[idx]) in ola/ola_arch.py.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start in ola/ola_arch.py at the mm_projector call shown in the issue, then trace the shapes of the low-resolution and high-resolution image features and the projector inputs for Ola-7b. Reproduce the reported RuntimeError and identify the source of the 2304-versus-1152 mismatch; done means the same evaluation path no longer raises this shape error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- computer-vision, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100