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

217 Commits (712edb043ac83aa7a8235dafdfa4fc299448d84e)
 

Author SHA1 Message Date
gensmusic 712edb043a tutorial build add CPPFLAGS which contains cuda include pathi when ./configure --with-cuda (#833) 2019-05-24 14:46:53 +02:00
Pierre Letessier 5d1ed5b6fb Fix unable to read untrained index (#817) 2019-05-07 18:14:15 +02:00
Lucas Hosseini 012954ebbf
Fix HNSW.fill_with_random_links. (#801) 2019-04-23 16:52:39 +02:00
Xuhui Sun 0a8f92d6d6 Update cpp tutorial makefile for GPU part (#774) 2019-04-11 14:20:38 +02:00
Matthijs Douze b460b0e9d5
Update INSTALL.md 2019-04-11 11:24:35 +02:00
Lucas Hosseini e811fecac6
Make clean target bourne shell compatible. (#783) 2019-04-10 15:54:57 +02:00
Lucas Hosseini 1e3e7f4262
Make Makefile bourne shell compliant. (#782) 2019-04-10 14:55:30 +02:00
Matthijs Douze f443db75f3 Merge branch 'master' of https://www.github.com/facebookresearch/faiss 2019-04-08 20:26:42 -07:00
Matthijs Douze fa91c13980 take into account torch offset when getting ptr 2019-04-08 20:26:26 -07:00
Steffen Schmitz d635aa9c65 fix: gpu compilation with cuda 10.1 (#771) 
The GPU compilation fails with g++ 8.3.1 and cuda 10.1
With this proposed fix the compilation works locally

Contributes to #751
 2019-04-08 20:39:21 +02:00
Lucas Hosseini 5555ae7f88
Configure install paths in Dockerfile. (#772) 2019-04-07 23:22:43 +02:00
Lucas Hosseini 7f5b22b0ff
Add conda packages metadata + tests. (#769) 
+ Add conda packages metadata (now building Faiss using conda's toolchain);
+ add Dockerfile for building conda packages (for all CUDA versions);
+ add working Dockerfile building faiss on Centos7;
+ simplify GPU build;
+ avoid falling back to CPU-only version (python);
+ simplify TravisCI config;
+ update INSTALL.md;
+ add configure flag for specifying target architectures (--with-cuda-arch);
+ fix Makefile for gpu tests;
+ fix various Makefile issues;
+ remove stale file (gpu/utils/DeviceUtils.cpp).
 2019-04-05 11:50:39 +02:00
Lucas Hosseini 6c1cb3cd5c Fix ivfpq demo. (#763) 
Makes sense.
 2019-04-03 11:24:33 +02:00
Lucas Hosseini afe0fdc161
Facebook sync (Mar 2019) (#756) 
Facebook sync (Mar 2019)

- MatrixStats object
- option to round coordinates during k-means optimization
- alternative option for search in HNSW
- moved stats and imbalance_factor of IndexIVF to InvertedLists object
- range search for IVFScalarQuantizer
- direct unit8 codec in ScalarQuantizer
- renamed IndexProxy to IndexReplicas and moved to main Faiss
- better support for PQ code assignment with external index
- support for IMI2x16 (4B virtual centroids!)
- support for k = 2048 search on GPU (instead of 1024)
- most CUDA mem alloc failures throw exceptions instead of terminating on an assertion
- support for renaming an ondisk invertedlists
- interrupt computations with ctrl-C in python
 2019-03-29 16:32:28 +01:00
Jeff Johnson a9959bf63b
Merge pull request #746 from facebookresearch/automated_fixup_code_of_conduct_file_exists 
Adding Code of Conduct file
 2019-03-25 11:43:16 -07:00
Facebook Community Bot 65d3d1f129 OSS Automated Fix: Addition of Code of Conduct 2019-03-22 11:13:52 -07:00
Lucas Hosseini 7b07685338
Update INSTALL.md 
Remove stale CMake reference.
 2019-03-06 12:12:03 +01:00
Marcin Tolysz 4b3210575b Improve arm64 support. (#676) 
* Improve Arm64 support.

Make library compile and run on Arm64 8.2 (Nvidia Xavier arm+cuda)
`configure` is regenerated

* remove simd from utils.cpp
 2019-02-15 13:13:49 +01:00
Arun-George-Zachariah b06cf4128d Upgrading pip in the Docker Container (#711) 2019-02-15 13:12:54 +01:00
Minjoon Seo c6a53c078b Bug fix: data needed to be added with ids (#723) 
When data is being added to each index block, it needs to explicitly identify ids.
 2019-02-10 05:44:05 +01:00
Lucas Hosseini 4e8d324922
Fix Travis build on OSX. (#692) 2019-01-15 21:47:02 +01:00
Matthijs Douze 702ad532db
Update README.md 2019-01-15 19:24:48 +01:00
Matthijs Douze fd0f0671d2
Update README.md 2019-01-15 19:22:16 +01:00
Corey J. Nolet f5feac149c For #669. Adding install target to gpu Makefile (#688) 2019-01-11 17:34:39 +01:00
Bonnie Varghese eb6c4cc96a Allow setting nprobe for IndexIDMap->IndexIVF (#673) 
There we go
 2019-01-09 13:00:36 +01:00
Lucas Hosseini f417a53628
Fix CI tests (#687) 
* Fix test_transfer_invlists.cpp

* Fix relative imports.

* Fix test_index_accuracy.py.

* Use default OSX version.

* Allow osx gcc6 build to fail.
 2019-01-08 17:52:36 +01:00
Eduardo Pinho 27a4ba0d93 [C API] Multi-GPU functions (#628) 
* [c_api] minor tweak on Index_c header imports

* [c_api] Multi GPU constructs

- Add constructor & destructor for `FaissGpuMultipleClonerOptions`
- Add function `faiss_index_cpu_to_gpu_multiple`
- Add function `faiss_index_cpu_to_gpu_multiple_with_options`

* [c_api] fix doc comments in GpuClonerOptions

* [c_api] Improve cpu_to_gpu_multiple prototype

* [c_api] Improve cpu_to_gpu_multiple prototype

* [c_api] Add dtor definition for GpuMultipleClonerOptions
 2018-12-23 18:46:07 +01:00
Lucas Hosseini ac7005b6ef Remove CMake. (#645) 2018-12-23 18:45:16 +01:00
Jeff Johnson aafc7b0457
Merge pull request #672 from irachex/patch-1 
Add missing `\` in gpu Makefile
 2018-12-22 19:36:05 -05:00
Huayi Zhang e23a80ebbf
Add missing `\` in gpu Makefile 2018-12-23 05:29:45 +08:00
Matthijs Douze 353b1967c2
Update README.md 2018-12-20 14:52:59 +01:00
Matthijs Douze 6a6bf40b2c
Create README.md 2018-12-20 14:45:46 +01:00
matthijs daf589d9d2 add bench_all_ivf 2018-12-20 05:43:36 -08:00
Lucas Hosseini 4bcb5b3fe0
Update README.md 2018-12-19 18:08:20 +01:00
Lucas Hosseini 323dbf3be3
Facebook sync (Dec 2018). (#660) 
* Add GpuIndexBinaryFlat
* Add IndexBinaryHNSW
 2018-12-19 17:48:35 +01:00
cyy 21475fb2f3 Use explicit numeric limits + add missing header (#644) 2018-11-23 14:29:58 +01:00
Ailing 7061699721 update INSTALL.md, added CUDA9.2, remove CUDA9.1 support (#583) 2018-11-23 14:23:17 +01:00
Alex Shen f56ea1c4de PYTHON executable location is missing in example makefile.inc (#626) 2018-11-23 14:18:33 +01:00
Matthijs Douze 87721af129
Update README.md 2018-10-03 14:09:51 +02:00
Tom Forbes a91a24e77a Fix #558 - Make `M` an integer (#589) 
* Fix #558 - Make `d` an integer

* Use int()
 2018-09-18 22:08:31 +02:00
Matthijs Douze e57270600a
Update README.md 2018-08-31 11:33:41 +02:00
Matthijs Douze cdc5a8b7b8
Update README.md 2018-08-30 21:15:45 +02:00
Matthijs Douze 6cd4cfde56
Update README.md 2018-08-30 21:15:14 +02:00
Lucas Hosseini 76bec0b500
Facebook sync (#573) 
Features:

- automatic tracking of C++ references in Python
- non-intel platforms supported -- some functions optimized for ARM
- override nprobe for concurrent searches
- support for floating-point quantizers in binary indexes
Bug fixes:

- no more segfaults in python (I know it's the same as the first feature but it's important!)
- fix GpuIndexIVFFlat issues for float32 with 64 / 128 dims
- fix sharding of flat indexes on GPU with index_cpu_to_gpu_multiple
 2018-08-30 19:38:50 +02:00
Ailing 19cea3d2f4 update conda install info for release 1.3.0 (#516) 2018-07-24 13:24:25 +02:00
Yohei Kikuta 67c6506bcc Replace os.tempnam with tempfile for python3 (#537) 
os.tempnam is eliminated in python3.
Used tempfile.NamedTemporaryFile().name instead.
This is also valid for python2.7.
 2018-07-24 10:25:40 +02:00
Lucas Hosseini 2dc30e14cf
Add support for Volta GPUs. (#517) 2018-07-10 19:37:57 +02:00
Lucas Hosseini a90c36a323
Update INSTALL.md 2018-07-06 20:23:25 +02:00
Lucas Hosseini 3c70207da5
Fix linking of python GPU faiss. (#510) 2018-07-06 17:31:25 +02:00
Lucas Hosseini 5d29b3624c
Use multi-threaded MKL by default (#509) 
* Configure script use threaded MKL instead of sequential.

* Do not link against GNU OpenMP when linking against Intel OpenMP.
 2018-07-06 17:26:37 +02:00