[−][src]Crate hnsw

Re-exports

pub use candidates::*;

pub use generic_array::typenum;

Euclidean	Any list, vector, etc of floats wrapped in `Euclidean` is to be treated as having euclidean distance.
GenericArray	Struct representing a generic array - `GenericArray<T, N>` works like [T; N]
HNSW	This provides a HNSW implementation for any distance function.
Hamming	Treats each bit contained in this struct as its own dimension and distance is computed as hamming distance.
Params
Searcher	Contains all the state used when searching the HNSW

ArrayLength	Trait making `GenericArray` work, marking types to be used as length of an array
Distance	This is the primary trait used by the HNSW. This is also implemented for `FloatingDistance`. If your features have a floating point distance, please implement the distance using `FloatingDistance`. Implementing `FloatingDistance` implements `Distance` so long as you satisfy its conditions.
FloatingDistance	Implement this trait when your features have a floating point distance between them. You will take no performance penalty for doing so. Please ensure your distance satisfies the conditions on `floating_distance`.