mirrors
/
faiss
mirror of https://github.com/facebookresearch/faiss.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
921 Commits
23 Branches
19 Tags
212 MiB
Commit Graph

2 Commits (9519a19f4291d5750c4b7c6036ef080522c7c299)

Author SHA1 Message Date
Denis Yaroshevskiy 45b16d23a1 faiss: use autovectorization for inner product (#2712) 
Summary:
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/2712

Using autovectorization to get the proper urnolling.

Previous version timings are:

Before
```
faiss_ip_10000                        2.10us   475.62K
faiss_n2_10000                        4.23us   236.30K
```

After
```
faiss_ip_10000                         1.21us   827.16K
faiss_n2_10000                       640.68ns     1.56M
```

Reviewed By: alexanderguzhva

Differential Revision: D43353199

fbshipit-source-id: 8f73a34acd4b0368be6cdb05ba7a99a566c9ed83
 2023-02-16 10:24:25 -08:00
Matthijs Douze b8fe92dfee contrib clustering module (#2217) 
Summary:
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/2217

This diff introduces a new Faiss contrib module that contains:
- generic k-means implemented in python (was in distributed_ondisk)
- the two-level clustering code, including a simple function that runs it on a Faiss IVF index.
- sparse clustering code (new)

The main idea is that that code is often re-used so better have it in contrib.

Reviewed By: beauby

Differential Revision: D34170932

fbshipit-source-id: cc297cc56d241b5ef421500ed410d8e2be0f1b77
 2022-02-28 14:18:47 -08:00