CosineSimilarity needs extra calculation, postgres does cosine distance

Cosine distance and cosine similarity are complementary measures. Cosine similarity measures how similar two vectors are, ranging from -1 (exactly opposite) to 1 (exactly the same). Cosine distance, on the other hand, measures how different two vectors are and is calculated as 1 minus the Cosine Similarity.

Issue created by @marcus_johansson
Merge request !2Resolve #3512198 "Cosinesimilarity needs extra" → (Merged) created by marcus_johansson
Comment about 1 month ago →
🇩🇪Germany marcus_johansson
Comment about 1 month ago →
🇩🇪Germany marcus_johansson
Sorry, not correct calculation
Comment about 1 month ago →
🇩🇪Germany marcus_johansson
Sorry, the calculation is correct. It seems to be giving back lower numbers then Milvus, I'm not sure why, but the calculation seems sound.
Comment about 1 month ago →
🇩🇪Germany marcus_johansson
Comment about 1 month ago →
joshhytr
Thanks @marcus_johansson, this looks great. I've added a minor query on the MR
Comment 30 days ago →
🇩🇪Germany marcus_johansson
Thanks @joshhytr - I have done a change in the MR, please recheck.
Comment 29 days ago →
System Message

joshhytr → committed 81a9f538 on 1.0.x authored by marcus_johansson →
Resolve #3512198 "Cosinesimilarity needs extra"
Comment 29 days ago →
joshhytr
Thanks for this @marcus_johansson!
Comment 15 days ago →
System Message
Automatically closed - issue fixed for 2 weeks with no activity.

CosineSimilarity needs extra calculation, postgres does cosine distance

Problem/Motivation

Proposed resolution

Merge Requests

!2CosineSimilarity needs extra calculation, postgres does cosine distance
Merged

Comments & Activities

CosineSimilarity needs extra calculation, postgres does cosine distance

Problem/Motivation

Proposed resolution

Merge Requests

!2CosineSimilarity needs extra calculation, postgres does cosine distanceMerged

Comments & Activities

!2CosineSimilarity needs extra calculation, postgres does cosine distance
Merged