Matthijs Douze 8fc3775472 building blocks for hybrid CPU / GPU search (#2638) ...

Summary:
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/2638

This diff is a more streamlined way of searching IVF indexes with precomputed clusters.
This will be used for experiments with hybrid CPU / GPU search.

Reviewed By: algoriddle

Differential Revision: D41301032

fbshipit-source-id: a1d645fd0f2bf806454dfd04971edc0a6200d20d

2023-01-12 13:34:44 -08:00

..

CMakeLists.txt

Add facilities for approximate evaluation of min-k distances via heap. Affects RQ / PRQ / RQ_LUT / PRQ_LUT (#2633)

2023-01-03 14:39:11 -08:00

common_faiss_tests.py

Add manifold check for size 0 (#1867)

2021-05-09 22:30:31 -07:00

test_approx_topk.cpp

Add facilities for approximate evaluation of min-k distances via heap. Affects RQ / PRQ / RQ_LUT / PRQ_LUT (#2633)

2023-01-03 14:39:11 -08:00

test_autotune.py

…

test_binary_factory.py

…

test_binary_flat.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_binary_hashindex.py

Add manifold check for size 0 (#1867)

2021-05-09 22:30:31 -07:00

test_binary_io.py

…

test_build_blocks.py

building blocks for hybrid CPU / GPU search (#2638)

2023-01-12 13:34:44 -08:00

test_clone.py

add clone for IndexRefine and others + tests (#2539)

2022-11-21 04:43:39 -08:00

test_clustering.py

Non-uniform quantization of vector norms (#2037)

2021-10-11 14:13:16 -07:00

test_contrib_with_scipy.py

contrib clustering module (#2217)

2022-02-28 14:18:47 -08:00

test_contrib.py

Implementation of big-batch IVF search (single machine) (#2567)

2022-12-09 08:53:13 -08:00

test_cppcontrib_sa_decode.cpp

Speedup tests for cppcontrib_sadecode kernels (#2620)

2022-12-12 11:59:05 -08:00

test_cppcontrib_uintreader.cpp

Additional C++ templates for fast sa_decode: more codecs for Index2LevelDecoder (#2443)

2022-09-05 10:04:32 -07:00

test_dealloc_invlists.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_documentation.py

Generate python docstrings from doxygen comments. (#1969)

2021-10-10 22:14:48 -07:00

test_doxygen_documentation.py

Generate python docstrings from doxygen comments. (#1969)

2021-10-10 22:14:48 -07:00

test_extra_distances.py

Add manifold check for size 0 (#1867)

2021-05-09 22:30:31 -07:00

test_factory.py

Fix sub-object ownership of python interface of IVFSpectralHash

2022-10-04 07:54:00 -07:00

test_fast_scan_ivf.py

Support for parallelization in IVFFastScan over both queries and probes (#2380)

2022-08-31 05:37:53 -07:00

test_fast_scan.py

CodePacker for non-contiguous code layouts (#2625)

2022-12-21 11:06:53 -08:00

test_index_accuracy.py

CodePacker for non-contiguous code layouts (#2625)

2022-12-21 11:06:53 -08:00

test_index_binary_from_float.py

…

test_index_binary.py

Fix packaging (#2121)

2021-12-07 13:12:30 -08:00

test_index_composite.py

building blocks for hybrid CPU / GPU search (#2638)

2023-01-12 13:34:44 -08:00

test_index.py

Fix test timeouts (#2618)

2022-12-12 09:04:43 -08:00

test_io.py

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_ivflib.py

building blocks for hybrid CPU / GPU search (#2638)

2023-01-12 13:34:44 -08:00

test_ivfpq_codec.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_ivfpq_indexing.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_local_search_quantizer.py

about a magnitude faster LSQ/PLSQ training and encoding (#2650)

2023-01-06 09:06:34 -08:00

test_lowlevel_ivf.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_mem_leak.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_merge_index.py

CodePacker for non-contiguous code layouts (#2625)

2022-12-21 11:06:53 -08:00

test_merge.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_meta_index.py

T132029385 support merge for IndexFlatCodes (#2488)

2022-09-23 07:19:21 -07:00

test_omp_threads_py.py

…

test_omp_threads.cpp

…

test_ondisk_ivf.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_oom_exception.py

…

test_pairs_decoding.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_params_override.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_partition.py

…

test_pq_encoding.cpp

add remove and merge features for IndexFastScan (#2497)

2022-10-11 04:14:29 -07:00

test_product_quantizer.py

Introduce transposed centroid table to speedup ProductQuantizer::compute_codes() (#2562)

2022-11-06 08:32:54 -08:00

test_referenced_objects.py

…

test_refine.py

Generalize DistanceComputer for flat indexes (#2255)

2022-03-20 23:43:33 -07:00

test_residual_quantizer.py

fail early if RCQ norms table would become too large

2022-09-23 11:16:12 -07:00

test_rowwise_minmax.py

IndexRowwiseMinMax (#2439)

2022-09-05 06:59:41 -07:00

test_search_params.py

Scan exactly max_codes elements (#2647)

2023-01-09 00:11:07 -08:00

test_simdlib.cpp

Speedup exhaustive_L2sqr_blas for AVX2, ARM NEON and AVX512 (#2568)

2022-11-14 17:01:52 -08:00

test_sliding_ivf.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_standalone_codec.py

Speedup exhaustive_L2sqr_blas for AVX2, ARM NEON and AVX512 (#2568)

2022-11-14 17:01:52 -08:00

test_threaded_index.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

test_transfer_invlists.cpp

Put idx_t in the faiss namespace (#2582)

2022-11-30 08:25:30 -08:00

torch_test_contrib.py

Automatic type conversions for Python API (#2274)

2022-03-30 05:42:08 -07:00