I've learned the Lsh algorithm recently. And I found your implementation. Quite useful to me!
But as far as I know, there are different Lsh families for different distance measurement. It seems the index method you use is random planes for cosine distance, am I right?
I'd like to ask you:
is it necessary or right to expand the index function for different distance measurement?
if it's right and necessary, will you or let me do it?
I've learned the Lsh algorithm recently. And I found your implementation. Quite useful to me! But as far as I know, there are different Lsh families for different distance measurement. It seems the index method you use is random planes for cosine distance, am I right? I'd like to ask you: