vqa_yn inference using my fine-tuned model

[wuzhaodongaipython]

Thank you for job! I think this job will push the progress for medical AI.

I find that your demo is about segmentation. If I set seg_enable=False I can just have the answer.

However, I want to find where is the text input. In the demo you just give question input so that I cannot find the text input.

Moreover, I have fine-tuned my model from your guide in LaMed/src/train/train.py. I want to know how to use this fined-tuned model to give the inference.

Hoping for your reply! I will cite this paper in the future.