v1.9.0

Added

• Add AVX-512 implementation for the distance and scalar quantizer functions. (#3853)
• Allow k and M suffixes in IVF indexes (#3812)
• add reconstruct support to additive quantizers (#3752)
• introduce options for reducing the overhead for a clustering procedure (#3731)
• Add hnsw search params for bounded queue option (#3748)
• ROCm support (#3462)
• Add sve targets (#2886)
• add get_version() for c_api (#3688)
• QINCo implementation in CPU Faiss (#3608)
• Add search functionality to FlatCodes (#3611)
• add dispatcher for VectorDistance and ResultHandlers (#3627)
• Add SQ8bit signed quantization (#3501)
• Add ABS_INNER_PRODUCT metric (#3524)
• Interop between CAGRA and HNSW (#3252)
• add skip_storage flag to HNSW (#3487)
• QT_bf16 for scalar quantizer for bfloat16 (#3444)
• Implement METRIC.NaNEuclidean (#3414)
• TimeoutCallback C++ and Python (#3417)
• support big-endian machines (#3361)
• Support for Remove ids from IVFPQFastScan index (#3354)
• Implement reconstruct_n for GPU IVFFlat indexes (#3338)
• Support of skip_ids in merge_from_multiple function of OnDiskInvertedLists (#3327)
• Add the ability to clone and read binary indexes to the C API. (#3318)
• AVX512 for PQFastScan (#3276)

Changed

• faster hnsw CPU index training (#3822)
• Some small improvements. (#3692)
• First attempt at LSH matching with nbits (#3679)
• Set verbosoe before train (#3619)
• Remove duplicate NegativeDistanceComputer instances (#3450)
• interrupt for NNDescent (#3432)
• Get rid of redundant instructions in ScalarQuantizer (#3430)
• PowerPC, improve code generation for function fvec_L2sqr (#3416)
• Unroll loop in lookup_2_lanes (#3364)
• Improve filtering & search parameters propagation (#3304)
• Change index_cpu_to_gpu to throw for indices not implemented on GPU (#3336)
• Throw when attempting to move IndexPQ to GPU (#3328)
• Skip HNSWPQ sdc init with new io flag (#3250)

Fixed

• FIx a bug for a non-simdlib code of ResidualQuantizer (#3868)
• assign_index should default to null (#3855)
• Fix an incorrectly counted the number of computed distances for HNSW (#3840)
• Add error for overflowing nbits during PQ construction (#3833)
• Fix radius search with HSNW and IP (#3698)
• fix algorithm of spreading vectors over shards (#3374)
• Fix IndexBinary.assign Python method (#3384)
• Few fixes in bench_fw to enable IndexFromCodec (#3383)
• Fix the endianness issue in AIX while running the benchmark. (#3345)
• Fix faiss swig build with version > 4.2.x (#3315)
• Fix problems when using 64-bit integers. (#3322)
• Fix IVFPQFastScan decode function (#3312)
• Handling FaissException in few destructors of ResultHandler.h (#3311)
• Fix HNSW stats (#3309)
• AIX compilation fix for io classes (#3275)

V1.8.0

Added

• Added a new conda package faiss-gpu-raft alongside faiss-cpu and faiss-gpu
• Integrated IVF-Flat and IVF-PQ implementations in faiss-gpu-raft from RAFT by Nvidia [thanks @cjnolet and @tarang-jain]
• Added a context parameter to InvertedLists and InvertedListsIterator
• Added Faiss on Rocksdb demo to showing how inverted lists can be persisted in a key-value store
• Introduced Offline IVF framework powered by Faiss big batch search
• Added SIMD NEON Optimization for QT_FP16 in Scalar Quantizer. [thanks @naveentatikonda]
• Generalized ResultHandler and supported range search for HNSW and FastScan
• Introduced avx512 optimization mode and FAISS_OPT_LEVEL env variable [thanks @alexanderguzhva]
• Added search parameters for IndexRefine::search() and IndexRefineFlat::search()
• Supported large two-level clustering
• Added support for Python 3.11 and 3.12
• Added support for CUDA 12

Changed

• Used the benchmark to find Pareto optimal indices. Intentionally limited to IVF(Flat|HNSW),PQ|SQ indices
• Splitted off RQ encoding steps to another file
• Supported better NaN handling
• HNSW speedup + Distance 4 points [thanks @alexanderguzhva]

Fixed

• Fixed DeviceVector reallocations in Faiss GPU
• Used efSearch from params if provided in HNSW search
• Fixed warp synchronous behavior in Faiss GPU CUDA 12

Full Changelog: v1.7.4...v1.8.0

v1.7.4

Added

• Added big batch IVF search for conducting efficient search with big batches of queries
• Checkpointing in big batch search support
• Precomputed centroids support
• Support for iterable inverted lists for eg. key value stores
• 64-bit indexing arithmetic support in FAISS GPU
• IndexIVFShards now handle IVF indexes with a common quantizer
• Jaccard distance support
• CodePacker for non-contiguous code layouts
• Approximate evaluation of top-k distances for ResidualQuantizer and IndexBinaryFlat
• Added support for 12-bit PQ / IVFPQ fine quantizer decoders for standalone vector codecs (faiss/cppcontrib)
• Conda packages for osx-arm64 (Apple M1) and linux-aarch64 (ARM64) architectures
• Support for Python 3.10

Removed

• CUDA 10 is no longer supported in precompiled packages
• Removed Python 3.7 support for precompiled packages
• Removed constraint for using fine quantizer with no greater than 8 bits for IVFPQ, for example, now it is possible to use IVF256,PQ10x12 for a CPU index

Changed

• Various performance optimizations for PQ / IVFPQ for AVX2 and ARM for training (fused distance+nearest kernel), search (faster kernels for distance_to_code() and scan_list_*()) and vector encoding
• A magnitude faster CPU code for LSQ/PLSQ training and vector encoding (reworked code)
• Performance improvements for Hamming Code computations for AVX2 and ARM (reworked code)
• Improved auto-vectorization support for IP and L2 distance computations (better handling of pragmas)
• Improved ResidualQuantizer vector encoding (pooling memory allocations, avoid r/w to a temporary buffer)

Fixed

• HSNW bug fixed which improves the recall rate! Special thanks to zh Wang @hhy3 for this.
• Faiss GPU IVF large query batch fix
• Faiss + Torch fixes, re-enable k = 2048
• Fix the number of distance computations to match max_codes parameter
• Fix decoding of large fast_scan blocks

v1.7.3

Added

• Sparse k-means routines and moved the generic kmeans to contrib
• FlatDistanceComputer for all FlatCodes indexes
• Support for fast accumulation of 4-bit LSQ and RQ
• Product additive quantization support
• Support per-query search parameters for many indexes + filtering by ids
• write_VectorTransform and read_vectorTransform were added to the public API (by @AbdelrahmanElmeniawy)
• Support for IDMap2 in index_factory by adding "IDMap2" to prefix or suffix of the input String (by @AbdelrahmanElmeniawy)
• Support for merging all IndexFlatCodes descendants (by @AbdelrahmanElmeniawy)
• Remove and merge features for IndexFastScan (by @AbdelrahmanElmeniawy)
• Performance improvements: 1) specialized the AVX2 pieces of code speeding up certain hotspots, 2) specialized kernels for vector codecs (this can be found in faiss/cppcontrib)

Fixed

• Fixed memory leak in OnDiskInvertedLists::do_mmap when the file is not closed (by @AbdelrahmanElmeniawy)
• LSH correctly throws error for metric types other than METRIC_L2 (by @AbdelrahmanElmeniawy)

Faiss 1.7.2

ADDED

• Support LSQ on GPU (by @KinglittleQ)
• Support for exact 1D kmeans (by @KinglittleQ)
• LUT-based search for additive quantizers
• Autogenerated Python docstrings from Doxygen comments

CHANGED

• Cleanup of index_factory parsing

Faiss 1.7.1

Prepare for release v1.7.1. (#1915)

Summary: Pull Request resolved: https://github.com/facebookresearch/faiss/pull/1915

Reviewed By: mdouze

Differential Revision: D28749880

Pulled By: beauby

fbshipit-source-id: 4dcb0a5eac8f84d26570a0fb6f3e357bc2f1151c

Faiss 1.7.0

Bump version to 1.7.0. (#1652)

Summary: Pull Request resolved: https://github.com/facebookresearch/faiss/pull/1652

Reviewed By: mdouze

Differential Revision: D26077948

Pulled By: beauby

fbshipit-source-id: 599ee61edd2425250948577cb55d145d9179ab25

Faiss 1.6.5

Fix version string in conda builds.

Summary: Currently, conda version strings are built from the latest git tag, which starts with the letter `v`. This confuses conda, which orders v1.6.5 before 1.6.3.

Reviewed By: LowikC

Differential Revision: D25151276

fbshipit-source-id: 7abfb547fee3468b26fedb6637a15e725755daf3

Faiss 1.6.4

Features

• Arbitrary dimensions per sub-quantizer now allowed for GpuIndexIVFPQ.
• Brute-force kNN on GPU (bfKnn) now accepts int32 indices.
• Faiss CPU now supports Windows. Conda packages are available from the nightly channel.

Faiss 1.6.3

Sync 20200323. (#1157)

* Sync 20200323.

* Bump version.

* Remove warning filter.