  faiss | |
  gpu | |
| detail | Type of a subspace of a tensor |
  SubTensor | A SubDim-rank slice of a parent Tensor |
 SubTensor< TensorType, 0, PtrTraits > | Specialization for a view of a single value (0-dimensional) |
| traits | |
| RestrictPtrTraits | |
| DefaultPtrTraits | |
| ToCPUCloner | |
| ToGpuCloner | |
| ToGpuClonerMultiple | |
| GpuClonerOptions | Set some options on how to copy to GPU |
| GpuMultipleClonerOptions | |
| GpuParameterSpace | Parameter space and setters for GPU indexes |
| GpuIndex | |
| IntToLong | |
| GpuIndexFlat | |
| GpuIndexFlatL2 | |
| GpuIndexFlatIP | |
| GpuIndexIVF | |
| GpuIndexIVFFlat | |
| GpuIndexIVFPQ | IVFPQ index for the GPU |
| GpuResources | |
| FlatIndex | Holder of GPU resources for a particular flat index |
| IVFBase | Base inverted list functionality for IVFFlat and IVFPQ |
| IVFFlat | |
| IVFFlatScan | The class that we use to provide scan specializations |
| IVFFlatScan<-1, L2, T > | |
| IVFFlatScan< 0, L2, T > | |
| IVFFlatScan< 64, L2, float > | |
| IVFFlatScan< 128, L2, float > | |
| IVFFlatScan< 256, L2, float > | |
| IVFPQ | Implementing class for IVFPQ on the GPU |
| Converter | |
| Converter< float > | |
| LoadCode32 | |
| LoadCode32< 1 > | |
| LoadCode32< 2 > | |
| LoadCode32< 3 > | |
| LoadCode32< 4 > | |
| LoadCode32< 8 > | |
| LoadCode32< 12 > | |
| LoadCode32< 16 > | |
| LoadCode32< 20 > | |
| LoadCode32< 24 > | |
| LoadCode32< 28 > | |
| LoadCode32< 32 > | |
| LoadCode32< 40 > | |
| LoadCode32< 48 > | |
| LoadCode32< 56 > | |
| LoadCode32< 64 > | |
| LoadCodeDistances | |
| IndexProxy | |
| IndexWrapper | |
| StandardGpuResources | |
| Comparator | |
| ConvertTo | |
| ConvertTo< float > | |
| ConvertTo< float2 > | |
| ConvertTo< float4 > | |
| DeviceMemoryReservation | |
| DeviceMemory | Manages temporary memory allocations on a GPU device |
| DeviceTensor | |
| DeviceScope | |
| CublasHandleScope | RAII object to manage a cublasHandle_t |
| CudaEvent | |
| DeviceVector | |
| HostTensor | |
| Limits | |
| Limits< float > | |
| Limits< int > | |
| Limits< Pair< K, V > > | |
| LoadStore | |
| Math | |
| Math< float2 > | |
| Math< float4 > | |
| CublasGemm | |
| CublasGemm< float > | |
| BlockMerge | |
| BlockMerge< NumThreads, K, V, N, L, Dir, Comp, true > | Merging lists smaller than a block |
| BlockMerge< NumThreads, K, V, N, L, Dir, Comp, false > | Merging lists larger than a block |
| BitonicMergeStep | |
| BitonicMergeStep< K, V, 1, Dir, Comp, Low, true > | |
| BitonicMergeStep< K, V, N, Dir, Comp, Low, true > | |
| BitonicMergeStep< K, V, N, Dir, Comp, true, false > | |
| BitonicMergeStep< K, V, N, Dir, Comp, false, false > | |
| BitonicSortStep | |
| BitonicSortStep< K, V, 1, Dir, Comp > | |
| NoTypeTensor | |
| Pair | A simple pair type for CUDA device usage |
| Sum | |
| Min | |
| Max | |
| SegmentedReduce | |
| FinalBlockMerge | |
| FinalBlockMerge< 1, NumThreads, K, V, NumWarpQ, Dir, Comp > | |
| FinalBlockMerge< 2, NumThreads, K, V, NumWarpQ, Dir, Comp > | |
| FinalBlockMerge< 4, NumThreads, K, V, NumWarpQ, Dir, Comp > | |
| FinalBlockMerge< 8, NumThreads, K, V, NumWarpQ, Dir, Comp > | |
| BlockSelect | |
| BlockSelect< K, V, Dir, Comp, 1, NumThreadQ, ThreadsPerBlock > | Specialization for k == 1 (NumWarpQ == 1) |
| WarpSelect | |
| WarpSelect< K, V, Dir, Comp, 1, NumThreadQ, ThreadsPerBlock > | Specialization for k == 1 (NumWarpQ == 1) |
| StackDeviceMemory | |
| Range | |
| Stack | |
| Tensor | Our tensor type |
| GpuResourcesThrustAllocator | Allocator for Thrust that comes out of a specified memory space |
| KernelTimer | Utility class for timing execution of a kernel |
| CpuTimer | CPU wallclock elapsed timer |
| TensorInfo | |
| TensorInfoOffset | |
| TensorInfoOffset< T,-1 > | |
| WorkerThread | |
| AutoTuneCriterion | |
| OneRecallAtRCriterion | |
| IntersectionCriterion | |
| OperatingPoint | |
| OperatingPoints | |
| ParameterRange | Possible values of a parameter, sorted from least to most expensive/accurate |
| ParameterSpace | |
| RangeSearchResult | |
| IDSelector | |
| IDSelectorRange | |
| IDSelectorBatch | |
| BufferList | |
| Buffer | |
| RangeSearchPartialResult | Entries in the buffers are split per query |
| QueryResult | Result structure for a single query |
| ClusteringParameters | |
| Clustering | |
| HammingComputer4 | |
| HammingComputer8 | |
| HammingComputer16 | |
| HammingComputer20 | |
| HammingComputer32 | |
| HammingComputer64 | |
| HammingComputerM8 | |
| HammingComputerM4 | |
| HammingComputer | |
| GenHammingComputer8 | |
| GenHammingComputer16 | |
| GenHammingComputer32 | |
| GenHammingComputerM8 | |
| CMax | |
| CMin | |
| HeapArray | |
| Index | |
| Cloner | |
| IndexFlat | |
| IndexFlatIP | |
| IndexFlatL2 | |
| IndexFlatL2BaseShift | |
| IndexRefineFlat | |
| IndexFlat1D | Optimized version for 1D "vectors" |
| IndexIVF | |
| IndexIVFFlat | |
| CodeCmp | |
| IndexIVFPQ | |
| IndexIVFPQStats | |
| IndexIVFPQR | |
| IndexIVFPQCompact | |
| IndexLSH | |
| ArgSort | |
| SortedArray | |
| SemiSortedArray | |
| MinSumK | |
| IndexPQ | |
| IndexPQStats | |
| MultiIndexQuantizer | |
| IndexIDMap | |
| IndexShards | |
| IndexSplitVectors | |
| Score3Computer | |
| IndirectSort | |
| RankingScore2 | |
| SimulatedAnnealingParameters | Parameters used for the simulated annealing method |
| PermutationObjective | Abstract class for the loss function |
| ReproduceDistancesObjective | |
| SimulatedAnnealingOptimizer | Simulated annealing optimization algorithm for permutations |
| PolysemousTraining | Optimizes the order of indices in a ProductQuantizer |
| ProductQuantizer | |
| NopDistanceCorrection | |
| BaseShiftDistanceCorrection | |
| RandomGenerator | Random generator that can be used in multithreaded contexts |
| VectorTransform | |
| LinearTransform | |
| RandomRotationMatrix | Randomly rotate a set of vectors |
| PCAMatrix | |
| OPQMatrix | |
| RemapDimensionsTransform | |
| IndexPreTransform | |
| Options | |
1.8.5
