Fine-tuning biobert for a better questioning answering

Hi all,

I saw that the “questioning answering” results of Biobert on my dataset aren’t good enough, so I want to fine-tune it. My dataset contains clinical medical files - which have been taken by a nurse or physician during the patient diagnosis/routine checkup.
My main question is, how many examples of Question-Answer pairs should I annotate for this fine-tuning?


hello did you find answers for your query? I was going to ask the same thing and I also have the same dataset structure as yours.