Hi, I just found that some answers in the original Dataset's json or mine cannot match the label size (i.e., answer_list) in your file.
Could you simply provide the total dataset size for medvqa-2019?
If you can provide the dataset you use, it will be very nice!
thank you for your works!
Hi, I just found that some answers in the original Dataset's json or mine cannot match the label size (i.e., answer_list) in your file.
Could you simply provide the total dataset size for medvqa-2019?
If you can provide the dataset you use, it will be very nice!
thank you for your works!