hnsw

type HNSW struct {
	sync.RWMutex
	// contains filtered or unexported fields
}

type Node struct {
	Connections [][]uint32 // Links to other nodes
	Vector      []float32  // Vector (X dimensions)
	Layer       int        // Layer the node exists in the HNSW tree
	ID          uint32     // Unique identifier
}

type Options struct {
	// M specifies the number of established connections for every new element during construction.
	// Reasonable range for M is 2-100. Higher M works better on datasets with high intrinsic dimensionality and/or high recall,
	// while low M works better for datasets with low intrinsic dimensionality and/or low recalls.
	// As an example, for dimension=4 random vectors, optimal M for search is somewhere around 6, while for high-dimensional datasets
	// (word embeddings, good face descriptors), higher M values are required (e.g., M=48-64) for optimal performance at high recall.
	// The range M=12-48 is ok for most use cases. When M is changed, one has to update the other parameters.
	// Nonetheless, ef and ef_construction parameters can be roughly estimated by assuming that M * ef_construction is a constant.
	M int

	// EF specifies the size of the dynamic candidate list.
	// EF is important for search time. Larger EF values can improve recall at the cost of increased search time.
	EF int

	// Heuristic indicates whether to use the heuristic algorithm (true) or the naive K-NN algorithm (false).
	// The heuristic algorithm is generally faster but may sacrifice some accuracy.
	Heuristic bool

	// DistanceType represents the type of distance function used for calculating distances between vectors.
	DistanceType index.DistanceType
}