You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This should come after implementing regular LSH for all of the similarities.
I don't think the multiprobe paradigm makes sense for Hamming and Jaccard. It might for L1 but I haven't done a lot of research on that. I know it makes sense for L2, since that's what the original paper did, and I'm fairly certain I've seen it described for cosine similarity.
The text was updated successfully, but these errors were encountered:
This should come after implementing regular LSH for all of the similarities.
I don't think the multiprobe paradigm makes sense for Hamming and Jaccard. It might for L1 but I haven't done a lot of research on that. I know it makes sense for L2, since that's what the original paper did, and I'm fairly certain I've seen it described for cosine similarity.
The text was updated successfully, but these errors were encountered: