manas6266

Regarding fine tuning for Custom VQA dataset

2

Thanks for your great work, I am interested in fine-tuning OFA for Visual Question Answering (VQA) using my custom dataset, which includes image-question-answer pairs. However, my dataset lacks confidence scores...

How to finetune the OFA model for grounded captioning? There is no clear description of it. Could you please tell it in detail??