open-compass / open-compass/VLMEvalKit
How is MMbench overall accuracy calculated?
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 4.4k
- Forks
- 768
- Avg merge
- 1d 10h
- Merged PRs (30d)
- 17
Description
I didn't find a description of the overall accuracy in the paper, i.e. is the overall accuracy the mean of the accuracy of each of the L2 categories, or per sample? Similar question for L2-L3.
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
The issue names MMbench overall accuracy and the L2/L3 categories but no file, test, or entry point. Start by locating the MMbench evaluation calculation and its existing documentation, then verify whether accuracy is aggregated by category or sample. Done means documenting the confirmed aggregation rules for overall, L2, and L3 accuracy.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100