MultiMedEval
MultiMedEval copied to clipboard
corentin-ryr
A Python tool to evaluate the performance of VLM on the medical domain.
Do you have any plans to evaluate more multimodal medical models? E.g. BiomedCLIP, PubMedCLIP, etc.?
Hello, thanks for your excellent work! I'm currently running VQA_RAD and PATH-VQA. Despite metrics have finished calculated, the progress bar fails to update. Could you please advise on how to...
Add ability to have stratified results when possible: - diverse demographics - types of diseases - sources of medical images Add ability to also take care of: fairness metrics such...
- Added BiomedParse segmentation tasks, including subtasks divided by modality. - Optimized metrics and logging for segmentation task. - Changed behavior of `BatcherOutput`
Thank you so much @corentin-ryr for such a great work, Could you please guide me to add more datasets for each task to evaluate VLMs on latest test datasets also?,...
Created KiTS Segmentation Task To-do: - [ ] Change BatcherInput images, segmentation masks to np.array. - [ ] Re-work the prompts. - [ ] Integration test with Segmentation from task_families.
Metadata
Owner
Metadata
A Python tool to evaluate the performance of VLM on the medical domain.