How can I compute the similarity scores
ClaudiaShu opened this issue · 1 comments
ClaudiaShu commented
Hi! thank you for your great work!
I've noticed that you used the cosine similarity function, are there any reasons for choosing this method?
mkhademi commented
Hi, no special reason other than cosign similarity was a simple, reasonable choice here. We just wanted to have a function to compare the embedding of different data points and cosine similarity was good enough for this purpose.