mirrors
/
faiss
mirror of https://github.com/facebookresearch/faiss.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
faiss/docs/html/search/all_73.js

93 lines
19 KiB
JavaScript
Raw Normal View History

Initial commit
2017-02-23 06:26:44 +08:00
var searchData=
[
Synchronization with FB version 2017-06-21 * moved most FAISS_ASSERT calls to C++ exceptions, and adjusted memory allocation to avoid mem leaks * added an IndexIVFScalarQuantizer type that offers an intermediate compression between IVFFlat and IVFPQ * support removal of indices in IndexIDMap / IndexFlat combination * various fixes in GPU code
2017-06-21 21:54:28 +08:00
['scalarquantizer',['ScalarQuantizer',['../structfaiss_1_1ScalarQuantizer.html',1,'faiss']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['scan_5fcodes',['scan_codes',['../structfaiss_1_1BinaryInvertedListScanner.html#a9ddb2bb0b513829b39528463c46f90f5',1,'faiss::BinaryInvertedListScanner::scan_codes()'],['../structfaiss_1_1InvertedListScanner.html#a218aeaf7cc45bf0bf27c3a8bfa44e303',1,'faiss::InvertedListScanner::scan_codes()']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['scan_5fcodes_5frange',['scan_codes_range',['../structfaiss_1_1InvertedListScanner.html#a57b6dcddab11cc801ce6288e065e1fa1',1,'faiss::InvertedListScanner']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['scan_5ftable_5fthreshold',['scan_table_threshold',['../structfaiss_1_1IVFPQSearchParameters.html#a5537033b0fe6cf3ef8334a16de7a47d5',1,'faiss::IVFPQSearchParameters::scan_table_threshold()'],['../structfaiss_1_1IndexIVFPQ.html#a7430fc4fa030c96497abced3b68358d4',1,'faiss::IndexIVFPQ::scan_table_threshold()']]],
['scopedcodes',['ScopedCodes',['../structfaiss_1_1InvertedLists_1_1ScopedCodes.html',1,'faiss::InvertedLists']]],
Synchronization with FB version 2017-06-21 * moved most FAISS_ASSERT calls to C++ exceptions, and adjusted memory allocation to avoid mem leaks * added an IndexIVFScalarQuantizer type that offers an intermediate compression between IVFFlat and IVFPQ * support removal of indices in IndexIDMap / IndexFlat combination * various fixes in GPU code
2017-06-21 21:54:28 +08:00
['scopedeleter',['ScopeDeleter',['../structfaiss_1_1ScopeDeleter.html',1,'faiss']]],
['scopedeleter1',['ScopeDeleter1',['../structfaiss_1_1ScopeDeleter1.html',1,'faiss']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['scopedids',['ScopedIds',['../structfaiss_1_1InvertedLists_1_1ScopedIds.html',1,'faiss::InvertedLists']]],
Synchronization with FB version 2017-06-21 * moved most FAISS_ASSERT calls to C++ exceptions, and adjusted memory allocation to avoid mem leaks * added an IndexIVFScalarQuantizer type that offers an intermediate compression between IVFFlat and IVFPQ * support removal of indices in IndexIDMap / IndexFlat combination * various fixes in GPU code
2017-06-21 21:54:28 +08:00
['scopefilecloser',['ScopeFileCloser',['../structfaiss_1_1ScopeFileCloser.html',1,'faiss']]],
Initial commit
2017-02-23 06:26:44 +08:00
['score3computer',['Score3Computer',['../structfaiss_1_1Score3Computer.html',1,'faiss']]],
['score3computer_3c_20float_2c_20double_20_3e',['Score3Computer< float, double >',['../structfaiss_1_1Score3Computer.html',1,'faiss']]],
['sdc_5ftable',['sdc_table',['../structfaiss_1_1ProductQuantizer.html#a6bd65fb4407452f73acb7242931d09e4',1,'faiss::ProductQuantizer']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['search',['search',['../classfaiss_1_1gpu_1_1GpuIndex.html#a4564ff946e4339156173b39449562971',1,'faiss::gpu::GpuIndex::search()'],['../classfaiss_1_1gpu_1_1GpuIndexBinaryFlat.html#aa4d81e853f4194a8b865fcfe7dddd050',1,'faiss::gpu::GpuIndexBinaryFlat::search()'],['../structfaiss_1_1HNSW.html#a330107a5801680609932849fc9404594',1,'faiss::HNSW::search()'],['../structfaiss_1_1Index.html#aced51b1ebc33c47ab3ae15ea906559a7',1,'faiss::Index::search()'],['../structfaiss_1_1IndexBinary.html#a48cdd06037d1fb8ce827e7522ba05fb8',1,'faiss::IndexBinary::search()'],['../structfaiss_1_1IndexBinaryFlat.html#a473e35ebabadc79869ca89f987553b13',1,'faiss::IndexBinaryFlat::search()'],['../structfaiss_1_1IndexBinaryFromFloat.html#af92913a782c65843fbf7a65e017d6f5f',1,'faiss::IndexBinaryFromFloat::search()'],['../structfaiss_1_1IndexBinaryHNSW.html#ae387b8e7094bfe584ef873c377a13af0',1,'faiss::IndexBinaryHNSW::search()'],['../structfaiss_1_1IndexBinaryIVF.html#aeaf24ee802fe8f43879c735c363b6fd5',1,'faiss::IndexBinaryIVF::search()'],['../structfaiss_1_1IndexFlat.html#a85510d70f4675197df9988651e4692c9',1,'faiss::IndexFlat::search()'],['../structfaiss_1_1IndexFlatL2BaseShift.html#a16e9bf1aec9d18b3ebbe78fea6bfecc0',1,'faiss::IndexFlatL2BaseShift::search()'],['../structfaiss_1_1IndexRefineFlat.html#ad04444de1e2c26c463cc52f3c076dd41',1,'faiss::IndexRefineFlat::search()'],['../structfaiss_1_1IndexFlat1D.html#a65736f2900865cd156faba4fcd260d05',1,'faiss::IndexFlat1D::search()'],['../structfaiss_1_1IndexHNSW.html#ae0ddf6ab4a54239a25d6277e21e8548e',1,'faiss::IndexHNSW::search()'],['../structfaiss_1_1IndexHNSW2Level.html#a50a087a17ce5a7a1fbcca0836959b4ec',1,'faiss::IndexHNSW2Level::search()'],['../structfaiss_1_1IndexIVF.html#ae2fd47b7d52603659e269aa8f6abb613',1,'faiss::IndexIVF::search()'],['../structfaiss_1_1Index2Layer.html#a5961efc2e19d9a6e5224f477b6d1b226',1,'faiss::Index2Layer::search()'],['../structfaiss_1_1IndexLSH.html#a66b6afb56afa1f58e2d93abfdfd223ee',1,'faiss::IndexLSH::search()'],['../structfaiss_1_1IndexPQ.html#a491ba3ed34093513e6a850fbca0abdd2',1,'faiss::IndexPQ::search()'],['../structfaiss_1_1MultiIndexQuantizer.html#a999ee03f4e65f453377e209d10e25681',1,'faiss::MultiIndexQuantizer::search()'],['../structfaiss_1_1MultiIndexQuantizer2.html#adff25d40c18f5d7a65d359d1cd80ea29',1,'faiss::MultiIndexQuantizer2::search()'],['../classfaiss_1_1IndexReplicasTemplate.html#ad6382e3ebaaca61607ad90fc80cdbd3c',1,'faiss::IndexReplicasTemplate::search()'],['../structfaiss_1_1IndexScalarQuantizer.html#abf24732c29106bb808148c3d30599662',1,'faiss::IndexScalarQuantizer::search()'],['../structfaiss_1_1IndexIDMap.html#a2f36fa9150a465b48952afbba856e2a1',1,'faiss::IndexIDMap::search()'],['../structfaiss_1_1IndexSplitVectors.html#a4a010daf51e44d9717e94400cc23c159',1,'faiss::IndexSplitVectors::search()'],['../structfaiss_1_1ProductQuantizer.html#a1ae7892a15b9f6809a3ba1a123a7831b',1,'faiss::ProductQuantizer::search()'],['../structfaiss_1_1IndexPreTransform.html#a7c2c820895d87083742969c281ae4911',1,'faiss::IndexPreTransform::search()']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['search_5fand_5freconstruct',['search_and_reconstruct',['../structfaiss_1_1Index.html#aa5b4e63e1064f9bf9c75625d114b353f',1,'faiss::Index::search_and_reconstruct()'],['../structfaiss_1_1IndexBinary.html#ae99e8745fc02ffac2cfdb431951eeabf',1,'faiss::IndexBinary::search_and_reconstruct()'],['../structfaiss_1_1IndexBinaryIVF.html#a75260a5d45fb24a9a3c3462786f1de4d',1,'faiss::IndexBinaryIVF::search_and_reconstruct()'],['../structfaiss_1_1IndexIVF.html#a77aec24920024a09f7661790d9534c10',1,'faiss::IndexIVF::search_and_reconstruct()'],['../structfaiss_1_1IndexPreTransform.html#a97ceaf3f87d3d729846762295351e972',1,'faiss::IndexPreTransform::search_and_reconstruct()']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['search_5fbounded_5fqueue',['search_bounded_queue',['../structfaiss_1_1HNSW.html#a683738e8ab8e67fcabd823510c30b074',1,'faiss::HNSW']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['search_5ffrom_5fcandidates',['search_from_candidates',['../structfaiss_1_1HNSW.html#ad91ac2e25c896dab7fb9d32f94b60a6a',1,'faiss::HNSW']]],
Initial commit
2017-02-23 06:26:44 +08:00
['search_5fip',['search_ip',['../structfaiss_1_1ProductQuantizer.html#a66be226806ca2abcaaf1653cdc690aa6',1,'faiss::ProductQuantizer']]],
sync with FB version 2018-01-09 - adding HNSW indexing method - simultaneous search and reconstruction for IndexIVFPQ
2018-01-09 22:44:43 +08:00
['search_5flevel_5f0',['search_level_0',['../structfaiss_1_1IndexHNSW.html#a5b619a41ddd52f0377efb701829c1e43',1,'faiss::IndexHNSW']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['search_5fpreassigned',['search_preassigned',['../structfaiss_1_1IndexBinaryIVF.html#a4f33ec7083279f26c0966c5445ec702d',1,'faiss::IndexBinaryIVF::search_preassigned()'],['../structfaiss_1_1IndexIVF.html#a7cf3a1963394f87d61913420a66eb468',1,'faiss::IndexIVF::search_preassigned()'],['../structfaiss_1_1IndexIVFFlatDedup.html#a88f84494a02e8ceab79982bcbcab4fa5',1,'faiss::IndexIVFFlatDedup::search_preassigned()'],['../structfaiss_1_1IndexIVFPQR.html#ab9daed04d3cb3f7f57f32c6b477ac5b3',1,'faiss::IndexIVFPQR::search_preassigned()']]],
Initial commit
2017-02-23 06:26:44 +08:00
['search_5ftype_5ft',['Search_type_t',['../structfaiss_1_1IndexPQ.html#a4fa05430935a02b62b7c15c9840c42fe',1,'faiss::IndexPQ']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['searchfromcpupaged_5f',['searchFromCpuPaged_',['../classfaiss_1_1gpu_1_1GpuIndexBinaryFlat.html#a126b2d8af97627f062deaf955164bb41',1,'faiss::gpu::GpuIndexBinaryFlat']]],
['searchimpl_5f',['searchImpl_',['../classfaiss_1_1gpu_1_1GpuIndex.html#a184c443785b715fd692b2b67aa71d687',1,'faiss::gpu::GpuIndex::searchImpl_()'],['../classfaiss_1_1gpu_1_1GpuIndexFlat.html#abe12febd18da31fb6c3355a2c6bff3d4',1,'faiss::gpu::GpuIndexFlat::searchImpl_()'],['../classfaiss_1_1gpu_1_1GpuIndexIVFFlat.html#abc85b8bc2e42b7d399f9c2561ae14ad0',1,'faiss::gpu::GpuIndexIVFFlat::searchImpl_()'],['../classfaiss_1_1gpu_1_1GpuIndexIVFPQ.html#a73a06501e87117e5f620bb07d7c1e54d',1,'faiss::gpu::GpuIndexIVFPQ::searchImpl_()']]],
['seed',['seed',['../structfaiss_1_1ClusteringParameters.html#a509c65e2ebe6ecabebd163ecb03c5579',1,'faiss::ClusteringParameters']]],
Initial commit
2017-02-23 06:26:44 +08:00
['segmentedreduce',['SegmentedReduce',['../structfaiss_1_1gpu_1_1SegmentedReduce.html',1,'faiss::gpu']]],
sync with FB version 2018-01-09 - adding HNSW indexing method - simultaneous search and reconstruction for IndexIVFPQ
2018-01-09 22:44:43 +08:00
['set',['set',['../structfaiss_1_1VisitedTable.html#a387b6428f4bfa44495c70a5f1f45f853',1,'faiss::VisitedTable']]],
['set_5fdefault_5fprobas',['set_default_probas',['../structfaiss_1_1HNSW.html#ae7d955690da5d6376c203e81e124ba94',1,'faiss::HNSW']]],
Initial commit
2017-02-23 06:26:44 +08:00
['set_5fderived_5fvalues',['set_derived_values',['../structfaiss_1_1ProductQuantizer.html#ab3067f333b26811537a8519fb08292ce',1,'faiss::ProductQuantizer']]],
['set_5fgroundtruth',['set_groundtruth',['../structfaiss_1_1AutoTuneCriterion.html#a174f0de8e6fd01905f5fa2170f5039d8',1,'faiss::AutoTuneCriterion']]],
['set_5findex_5fparameter',['set_index_parameter',['../structfaiss_1_1ParameterSpace.html#ac630c153ffd12f9994452c92fe57dc17',1,'faiss::ParameterSpace::set_index_parameter()'],['../structfaiss_1_1gpu_1_1GpuParameterSpace.html#a1820826bb47a51ec9f000118be9c83b2',1,'faiss::gpu::GpuParameterSpace::set_index_parameter()']]],
['set_5findex_5fparameters',['set_index_parameters',['../structfaiss_1_1ParameterSpace.html#a55783994cee46e9181bad2b3a7a973f4',1,'faiss::ParameterSpace::set_index_parameters(Index *index, size_t cno) const '],['../structfaiss_1_1ParameterSpace.html#aef1f2af781a71e7f0cf6f67603ef7cc1',1,'faiss::ParameterSpace::set_index_parameters(Index *index, const char *param_string) const ']]],
sync with FB version 2018-01-09 - adding HNSW indexing method - simultaneous search and reconstruction for IndexIVFPQ
2018-01-09 22:44:43 +08:00
['set_5fis_5forthonormal',['set_is_orthonormal',['../structfaiss_1_1LinearTransform.html#aca77897ef5c27c69374b4d9555e5b332',1,'faiss::LinearTransform']]],
Initial commit
2017-02-23 06:26:44 +08:00
['set_5flims',['set_lims',['../structfaiss_1_1RangeSearchPartialResult.html#ab6a385008493c7b630e4e7bc56e3da74',1,'faiss::RangeSearchPartialResult']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['set_5flist',['set_list',['../structfaiss_1_1BinaryInvertedListScanner.html#ae28e1381cb4c16ff9941c909fe9c4122',1,'faiss::BinaryInvertedListScanner::set_list()'],['../structfaiss_1_1InvertedListScanner.html#ae801ea7ba0086a05ace3b779c3acf305',1,'faiss::InvertedListScanner::set_list()']]],
sync with FB version 2018-01-09 - adding HNSW indexing method - simultaneous search and reconstruction for IndexIVFPQ
2018-01-09 22:44:43 +08:00
['set_5fnb_5fneighbors',['set_nb_neighbors',['../structfaiss_1_1HNSW.html#afdb5a17487296267efaaa94e6a08996d',1,'faiss::HNSW']]],
Initial commit
2017-02-23 06:26:44 +08:00
['set_5fparams',['set_params',['../structfaiss_1_1ProductQuantizer.html#ad291f113c6d44a634ddc1972f9a3c775',1,'faiss::ProductQuantizer']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['set_5fquery',['set_query',['../structfaiss_1_1DistanceComputer.html#aaa4472453365b786adc2ab21e9cee5ec',1,'faiss::DistanceComputer::set_query()'],['../structfaiss_1_1BinaryInvertedListScanner.html#a22bf6ac0128e4e45768f91b4de58fdb6',1,'faiss::BinaryInvertedListScanner::set_query()'],['../structfaiss_1_1InvertedListScanner.html#a97cab8fbf8203bdda6c42fc4b8f92319',1,'faiss::InvertedListScanner::set_query()']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['setcudamallocwarning',['setCudaMallocWarning',['../classfaiss_1_1gpu_1_1StandardGpuResources.html#a3caad981d50370b9f5c121509eb2ed17',1,'faiss::gpu::StandardGpuResources::setCudaMallocWarning()'],['../classfaiss_1_1gpu_1_1StackDeviceMemory.html#aadabeef74940ef0cb6a965752bbf6156',1,'faiss::gpu::StackDeviceMemory::setCudaMallocWarning()']]],
Sync with FB version 2017-07-16 - access GPU resources - all Hamming distances of all sizes supported
2017-07-26 18:11:01 +08:00
['setdefaultnullstreamalldevices',['setDefaultNullStreamAllDevices',['../classfaiss_1_1gpu_1_1StandardGpuResources.html#a5374f336f1957c266d7a5c185eb962a1',1,'faiss::gpu::StandardGpuResources']]],
['setdefaultstream',['setDefaultStream',['../classfaiss_1_1gpu_1_1StandardGpuResources.html#a5924b424c36dba8300866f1b98c438cb',1,'faiss::gpu::StandardGpuResources']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['setminpagingsize',['setMinPagingSize',['../classfaiss_1_1gpu_1_1GpuIndex.html#a6c82cbea6b02cc996d634f822c1d310b',1,'faiss::gpu::GpuIndex']]],
Initial commit
2017-02-23 06:26:44 +08:00
['setnumprobes',['setNumProbes',['../classfaiss_1_1gpu_1_1GpuIndexIVF.html#a1efac4b9801d57fa6d0085d6aed9c1f2',1,'faiss::gpu::GpuIndexIVF']]],
['setpinnedmemory',['setPinnedMemory',['../classfaiss_1_1gpu_1_1StandardGpuResources.html#af9acb23ac323f85fa5dab4ed968ccd51',1,'faiss::gpu::StandardGpuResources']]],
['setprecomputedcodes',['setPrecomputedCodes',['../classfaiss_1_1gpu_1_1GpuIndexIVFPQ.html#afda2894c0ce520e0d3d06eeca5ed5f0c',1,'faiss::gpu::GpuIndexIVFPQ::setPrecomputedCodes()'],['../classfaiss_1_1gpu_1_1IVFPQ.html#adcee5dbf48c3cb6b8a67f5f392e155fd',1,'faiss::gpu::IVFPQ::setPrecomputedCodes()']]],
['settempmemory',['setTempMemory',['../classfaiss_1_1gpu_1_1StandardGpuResources.html#a367102e65b16bfdc7a32770a4f5854a6',1,'faiss::gpu::StandardGpuResources']]],
['shard',['shard',['../structfaiss_1_1gpu_1_1GpuMultipleClonerOptions.html#a7c5bd7a840f6641d5d68699f4416d79d',1,'faiss::gpu::GpuMultipleClonerOptions']]],
sync with FB version 2017-11-22 various bugfixes from github issues kmean with some frozen centroids GPU better tiling for large flat datasets default AVX for vector ops
2017-11-22 21:00:42 +08:00
['shard_5ftype',['shard_type',['../structfaiss_1_1gpu_1_1GpuMultipleClonerOptions.html#a9548f823297298b0a78109dec7feb45e',1,'faiss::gpu::GpuMultipleClonerOptions']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['shrink_5fneighbor_5flist',['shrink_neighbor_list',['../structfaiss_1_1HNSW.html#a0cf1870a859783925e95b15bf5e83584',1,'faiss::HNSW']]],
sync with FB version 2018-02-23 (#347) - support on-disk IVF
2018-02-23 23:44:31 +08:00
['simulatedannealingoptimizer',['SimulatedAnnealingOptimizer',['../structfaiss_1_1SimulatedAnnealingOptimizer.html',1,'faiss']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['simulatedannealingoptimizer',['SimulatedAnnealingOptimizer',['../structfaiss_1_1SimulatedAnnealingOptimizer.html#ab258eaf892742e37dd2c3bfd3b1c39e1',1,'faiss::SimulatedAnnealingOptimizer']]],
Initial commit
2017-02-23 06:26:44 +08:00
['simulatedannealingparameters',['SimulatedAnnealingParameters',['../structfaiss_1_1SimulatedAnnealingParameters.html',1,'faiss']]],
sync with FB version 2017-11-22 various bugfixes from github issues kmean with some frozen centroids GPU better tiling for large flat datasets default AVX for vector ops
2017-11-22 21:00:42 +08:00
['size_5f',['size_',['../structfaiss_1_1gpu_1_1StackDeviceMemory_1_1Stack.html#a3df7ba513acd46aa69180585cb4e7aac',1,'faiss::gpu::StackDeviceMemory::Stack::size_()'],['../classfaiss_1_1gpu_1_1Tensor.html#ad96fbf0f5e7c06a1031b8b18f7fc01d7',1,'faiss::gpu::Tensor::size_()']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['sizes',['sizes',['../structfaiss_1_1ivflib_1_1SlidingIndexWindow.html#a6dfc430601d194c428c2a25e3afa4492',1,'faiss::ivflib::SlidingIndexWindow::sizes()'],['../classfaiss_1_1gpu_1_1Tensor.html#abc0ecc4f882ee09632b5a06be0619adb',1,'faiss::gpu::Tensor::sizes()']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['sliceinvertedlists',['SliceInvertedLists',['../structfaiss_1_1SliceInvertedLists.html',1,'faiss']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['slidingindexwindow',['SlidingIndexWindow',['../structfaiss_1_1ivflib_1_1SlidingIndexWindow.html#a6347809123313ca945defdc7ddcd1465',1,'faiss::ivflib::SlidingIndexWindow']]],
['slidingindexwindow',['SlidingIndexWindow',['../structfaiss_1_1ivflib_1_1SlidingIndexWindow.html',1,'faiss::ivflib']]],
sync with FB version 2018-02-23 (#347) - support on-disk IVF
2018-02-23 23:44:31 +08:00
['slot',['Slot',['../structfaiss_1_1OnDiskInvertedLists_1_1Slot.html',1,'faiss::OnDiskInvertedLists']]],
Initial commit
2017-02-23 06:26:44 +08:00
['source_5fdis',['source_dis',['../structfaiss_1_1ReproduceDistancesObjective.html#a6681846eb7751b32a907d075fff6770b',1,'faiss::ReproduceDistancesObjective']]],
Synchronization with FB version 2017-06-21 * moved most FAISS_ASSERT calls to C++ exceptions, and adjusted memory allocation to avoid mem leaks * added an IndexIVFScalarQuantizer type that offers an intermediate compression between IVFFlat and IVFPQ * support removal of indices in IndexIDMap / IndexFlat combination * various fixes in GPU code
2017-06-21 21:54:28 +08:00
['space_5f',['space_',['../classfaiss_1_1gpu_1_1IVFBase.html#a30499eb61763df72cab25ea9d1235251',1,'faiss::gpu::IVFBase']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['spherical',['spherical',['../structfaiss_1_1ClusteringParameters.html#ad997fb511f574f7ddc69938c21612f8d',1,'faiss::ClusteringParameters']]],
sync with FB version 2017-07-18 - implemented ScalarQuantizer (without IVF) - implemented update for IndexIVFFlat - implemented L2 normalization preproc
2017-07-18 17:43:35 +08:00
['sq',['sq',['../structfaiss_1_1IndexScalarQuantizer.html#a71381f74754d8a54812e80af46a2a83a',1,'faiss::IndexScalarQuantizer']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['sqdistancecomputer',['SQDistanceComputer',['../structfaiss_1_1SQDistanceComputer.html',1,'faiss']]],
Initial commit
2017-02-23 06:26:44 +08:00
['st_5fgeneralized_5fhe',['ST_generalized_HE',['../structfaiss_1_1IndexPQ.html#a4fa05430935a02b62b7c15c9840c42fea237afca36629dae511c0ac56effbfaa6',1,'faiss::IndexPQ']]],
['st_5fhe',['ST_HE',['../structfaiss_1_1IndexPQ.html#a4fa05430935a02b62b7c15c9840c42fea903849f87cca2e7fbbfe034edbb81718',1,'faiss::IndexPQ']]],
['st_5fpolysemous',['ST_polysemous',['../structfaiss_1_1IndexPQ.html#a4fa05430935a02b62b7c15c9840c42feaf06a0e6b53f9cbe1c791c6912169b738',1,'faiss::IndexPQ']]],
['st_5fpolysemous_5fgeneralize',['ST_polysemous_generalize',['../structfaiss_1_1IndexPQ.html#a4fa05430935a02b62b7c15c9840c42feaef79af61798670a362434b7bc9ced45e',1,'faiss::IndexPQ']]],
['st_5fpq',['ST_PQ',['../structfaiss_1_1IndexPQ.html#a4fa05430935a02b62b7c15c9840c42fea00f77a4f8209c30c3ef37ccfd3d0c10a',1,'faiss::IndexPQ']]],
['st_5fsdc',['ST_SDC',['../structfaiss_1_1IndexPQ.html#a4fa05430935a02b62b7c15c9840c42fea4287cbc83d1d5f0d1a02f014e5fe692c',1,'faiss::IndexPQ']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['stack',['Stack',['../structfaiss_1_1gpu_1_1StackDeviceMemory_1_1Stack.html',1,'faiss::gpu::StackDeviceMemory']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['stack',['Stack',['../structfaiss_1_1gpu_1_1StackDeviceMemory_1_1Stack.html#a783de9fc71cf57bf336c474f6155453d',1,'faiss::gpu::StackDeviceMemory::Stack::Stack(int device, size_t size)'],['../structfaiss_1_1gpu_1_1StackDeviceMemory_1_1Stack.html#a92a16b44da0ab2a89d3d032b762c0232',1,'faiss::gpu::StackDeviceMemory::Stack::Stack(int device, void *p, size_t size, bool isOwner)']]],
Initial commit
2017-02-23 06:26:44 +08:00
['stack_5f',['stack_',['../classfaiss_1_1gpu_1_1StackDeviceMemory.html#a323a632f397eaf3df3034595762c6a29',1,'faiss::gpu::StackDeviceMemory']]],
sync with FB version 2017-09-08 fix index_cpu_to_gpu_multiple fix PCA dimension
2017-09-08 19:39:41 +08:00
['stackdevicememory',['StackDeviceMemory',['../classfaiss_1_1gpu_1_1StackDeviceMemory.html#a130f900c32ba91ae1f872c7ee506a93e',1,'faiss::gpu::StackDeviceMemory::StackDeviceMemory(int device, size_t allocPerDevice)'],['../classfaiss_1_1gpu_1_1StackDeviceMemory.html#afd7a84b8cd1f747d873cdd1ab07e8a4b',1,'faiss::gpu::StackDeviceMemory::StackDeviceMemory(int device, void *p, size_t size, bool isOwner)']]],
sync with FB version 2018-02-23 (#347) - support on-disk IVF
2018-02-23 23:44:31 +08:00
['stackdevicememory',['StackDeviceMemory',['../classfaiss_1_1gpu_1_1StackDeviceMemory.html',1,'faiss::gpu']]],
Initial commit
2017-02-23 06:26:44 +08:00
['standardgpuresources',['StandardGpuResources',['../classfaiss_1_1gpu_1_1StandardGpuResources.html',1,'faiss::gpu']]],
['start_5f',['start_',['../structfaiss_1_1gpu_1_1StackDeviceMemory_1_1Stack.html#aed64654e7d498114fdd48a6f1cdb1aa2',1,'faiss::gpu::StackDeviceMemory::Stack']]],
Facebook sync (Dec 2018). (#660) * Add GpuIndexBinaryFlat * Add IndexBinaryHNSW
2018-12-20 00:48:35 +08:00
['step',['step',['../structfaiss_1_1ivflib_1_1SlidingIndexWindow.html#accf1c7b8de188b74de0d6827beceac7c',1,'faiss::ivflib::SlidingIndexWindow']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['stop',['stop',['../classfaiss_1_1WorkerThread.html#a5230264fa557f6d77c5fed5601a8c857',1,'faiss::WorkerThread']]],
sync with FB version 2018-01-09 - adding HNSW indexing method - simultaneous search and reconstruction for IndexIVFPQ
2018-01-09 22:44:43 +08:00
['storage_5fidx_5ft',['storage_idx_t',['../structfaiss_1_1HNSW.html#ae1342fc2f07c325598a73b733e88ee74',1,'faiss::HNSW']]],
Synchronization with FB version 2017-06-21 * moved most FAISS_ASSERT calls to C++ exceptions, and adjusted memory allocation to avoid mem leaks * added an IndexIVFScalarQuantizer type that offers an intermediate compression between IVFFlat and IVFPQ * support removal of indices in IndexIDMap / IndexFlat combination * various fixes in GPU code
2017-06-21 21:54:28 +08:00
['storetransposed',['storeTransposed',['../structfaiss_1_1gpu_1_1GpuClonerOptions.html#a5f2470ec5b2a8fe7aae0542f113b781d',1,'faiss::gpu::GpuClonerOptions::storeTransposed()'],['../structfaiss_1_1gpu_1_1GpuIndexFlatConfig.html#a0342a26f5e59563bfe04f3f80e3b27f0',1,'faiss::gpu::GpuIndexFlatConfig::storeTransposed()']]],
Initial commit
2017-02-23 06:26:44 +08:00
['streamwaitonevent',['streamWaitOnEvent',['../classfaiss_1_1gpu_1_1CudaEvent.html#ad8968b298b95788c7679d7e5aa0571eb',1,'faiss::gpu::CudaEvent']]],
sync with FB version 2017-11-22 various bugfixes from github issues kmean with some frozen centroids GPU better tiling for large flat datasets default AVX for vector ops
2017-11-22 21:00:42 +08:00
['stride_5f',['stride_',['../classfaiss_1_1gpu_1_1Tensor.html#af4b8fe4b632cdca51ee7972ed93fc3fa',1,'faiss::gpu::Tensor']]],
['strides',['strides',['../classfaiss_1_1gpu_1_1Tensor.html#a87a777247486756e99060547a3cc833a',1,'faiss::gpu::Tensor']]],
Initial commit
2017-02-23 06:26:44 +08:00
['subtensor',['SubTensor',['../classfaiss_1_1gpu_1_1detail_1_1SubTensor.html',1,'faiss::gpu::detail']]],
['subtensor_3c_20tensortype_2c_200_2c_20ptrtraits_20_3e',['SubTensor< TensorType, 0, PtrTraits >',['../classfaiss_1_1gpu_1_1detail_1_1SubTensor_3_01TensorType_00_010_00_01PtrTraits_01_4.html',1,'faiss::gpu::detail']]],
['subtensor_3c_20tensortype_2c_201_2c_20ptrtraits_20_3e',['SubTensor< TensorType, 1, PtrTraits >',['../classfaiss_1_1gpu_1_1detail_1_1SubTensor_3_01TensorType_00_010_00_01PtrTraits_01_4.html#a5aa8367d35e6c281c29855d7cf24bd6d',1,'faiss::gpu::detail::SubTensor< TensorType, 0, PtrTraits >']]],
['subtensor_3c_20tensortype_2c_20subdim_2b1_2c_20ptrtraits_20_3e',['SubTensor< TensorType, SubDim+1, PtrTraits >',['../classfaiss_1_1gpu_1_1detail_1_1SubTensor.html#a33929e7387099e4e49be139ba467ebfc',1,'faiss::gpu::detail::SubTensor']]],
Sync with FB version 2017-07-16 - access GPU resources - all Hamming distances of all sizes supported
2017-07-26 18:11:01 +08:00
['sum',['Sum',['../structfaiss_1_1gpu_1_1Sum.html',1,'faiss::gpu']]],
Facebook sync (May 2019) + relicense (#838) Changelog: - changed license: BSD+Patents -> MIT - propagates exceptions raised in sub-indexes of IndexShards and IndexReplicas - support for searching several inverted lists in parallel (parallel_mode != 0) - better support for PQ codes where nbit != 8 or 16 - IVFSpectralHash implementation: spectral hash codes inside an IVF - 6-bit per component scalar quantizer (4 and 8 bit were already supported) - combinations of inverted lists: HStackInvertedLists and VStackInvertedLists - configurable number of threads for OnDiskInvertedLists prefetching (including 0=no prefetch) - more test and demo code compatible with Python 3 (print with parentheses) - refactored benchmark code: data loading is now in a single file
2019-05-28 22:17:22 +08:00
['symmetric_5fdis',['symmetric_dis',['../structfaiss_1_1DistanceComputer.html#a4b7236a757c788401aac00598ed902c7',1,'faiss::DistanceComputer']]],
Sync with FB version 2017-07-16 - access GPU resources - all Hamming distances of all sizes supported
2017-07-26 18:11:01 +08:00
['syncdefaultstream',['syncDefaultStream',['../classfaiss_1_1gpu_1_1GpuResources.html#a130deab9f6d49cbdde8f9da45ce868fd',1,'faiss::gpu::GpuResources']]],
['syncdefaultstreamcurrentdevice',['syncDefaultStreamCurrentDevice',['../classfaiss_1_1gpu_1_1GpuResources.html#adf6a5653ec2bbb0e9d58a0e1dd73897f',1,'faiss::gpu::GpuResources']]]
Initial commit
2017-02-23 06:26:44 +08:00
];