git clone https://github.com/wenkesj/rphashor
go get github.com/wenkesj/rphashsh rphash/installcd rphash/tests
go test -v -bench=.The official documentation for the high performance big data clustering algorithm RPHash.
An instance of the RPHashObject is the SimpleArray struct.
import "github.com/wenkesj/rphash/reader/simplearray"type SimpleArray struct {
data types.Iterator;
dimension int;
numberOfProjections int;
decoderMultiplier int;
randomSeed int64;
hashModulus int64;
k int;
numberOfBlurs int;
decoder types.Decoder;
centroids [][]float64;
topIDs []int64;
};func NewSimpleArray(X [][]float64, k int) *SimpleArrayReturns a new RPHashObject.
import "github.com/wenkesj/rphash/simple"type Simple struct {
centroids [][]float64
variance float64
rphashObject RPHashObject
}func NewSimple(_rphashObject RPHashObject) *SimpleNewSimple returns an instance of the Simple struct.
func (this *Simple) Map() RPHashObjectMaps all the default tasks to the RPHashObject. This will update and return the new RPHashObject.
func (this *Simple) Reduce() RPHashObjectPerforms all the default tasks on the RPHashObject. Updates and returns new RPHashObject.
func (this *Simple) GetCentroids() [][]float64Performs a KMeans operation on the Simple's centroids with the RPHashObject K value. Returns the calculated centroids.
func (this *Simple) Run()Performs the Map and Reduce functions and updates the centroids.
func (this *Simple) GetParam() RPHashObjectReturns the RPHashObject of the Simple struct.
import "github.com/wenkesj/rphash/stream"type Stream struct {
counts []int64;
centroids [][]float64;
variance float64;
centroidCounter types.CentroidItemSet;
random *rand.Rand;
rphashObject types.RPHashObject;
lshGroup []types.LSH;
decoder types.Decoder;
projector types.Projector;
hash types.Hash;
varTracker types.StatTest;
};func NewStream(_rphashObject types.RPHashObject) *Streamfunc (this *Stream) AddVectorOnlineStep(vec []float64) int64func (this *Stream) GetCentroids() [][]float64func (this *Stream) GetCentroidsOfflineStep() [][]float64func (this *Stream) Run()