Package org.redisson.api.search.index
Interface SVSVamanaVectorOptionalArgs
- All Superinterfaces:
FieldIndex
- All Known Implementing Classes:
SVSVamanaVectorIndexParams
Optional options object for vector field index which uses SVS-VAMANA indexing method.
- Author:
- seakider
-
Nested Class Summary
Nested ClassesModifier and TypeInterfaceDescriptionstatic enum -
Method Summary
Modifier and TypeMethodDescriptionDefines compression algorithmconstructionWindowSize(int value) Defines number of the search window size to during graph constructionepsilon(double value) Defines relative factor that sets the boundaries in which a range query may search for candidatesgraphMaxDegree(int value) Defines the maximum number of edges per nodeleanVecDim(int value) Defines the dimension used when using LeanVec4x8 or LeanVec8x8searchWindowSize(int value) Defines the size of the search windowtrainingThreshold(int value) Defines number of vectors needed to learn compression parameters.
-
Method Details
-
compression
Defines compression algorithm- Parameters:
algorithm- compression algorithm- Returns:
- vector options
-
constructionWindowSize
Defines number of the search window size to during graph construction- Parameters:
value- number of search window size- Returns:
- vector options
-
graphMaxDegree
Defines the maximum number of edges per node- Parameters:
value- the maximum number of edges- Returns:
- vector options
-
searchWindowSize
Defines the size of the search window- Parameters:
value- size of the search window- Returns:
- vector options
-
epsilon
Defines relative factor that sets the boundaries in which a range query may search for candidates- Parameters:
value- relative factor- Returns:
- vector options
-
trainingThreshold
Defines number of vectors needed to learn compression parameters.- Parameters:
value- number of vectors- Returns:
-
leanVecDim
Defines the dimension used when using LeanVec4x8 or LeanVec8x8- Parameters:
value- the dimension- Returns:
-