@Reference(type=Inproceedings, author={"Lv, Qin","Charikar, Moses","Li, Kai"}, title="Image similarity search with compact data structures", year="2004", booktitle="Proceedings of the thirteenth ACM international conference on Information and knowledge management", pages={"208","","217"}, url="http://doi.acm.org/10.1145/1031171.1031213", publisher="ACM", series="CIKM \'04") public class FloatHyperplaneL1Factory extends FloatHashFunctionFactory
The hash function hashes the input vector into a binary value (i.e. 0 or 1). It works by choosing a random dimension and a random threshold along that dimension (between a given minimum and maximum which define the closed space). Input vectors whose element at the chosen dimension is greater than or equal to the threshold generate a 1; values less than the threshold generate a 0.
ndims, rng
Constructor and Description |
---|
FloatHyperplaneL1Factory(int ndims,
cern.jet.random.engine.MersenneTwister rng,
float min,
float max)
Construct with the given arguments.
|
Modifier and Type | Method and Description |
---|---|
org.openimaj.lsh.functions.FloatHyperplaneL1Factory.Function |
create()
Construct a new
HashFunction . |
protected FloatFVComparison |
fvDistanceFunction() |
distanceFunction
public FloatHyperplaneL1Factory(int ndims, cern.jet.random.engine.MersenneTwister rng, float min, float max)
ndims
- The number of dimensionsrng
- A random number generatormin
- The minimum bound of the spacemax
- The maximum bound of the spacepublic org.openimaj.lsh.functions.FloatHyperplaneL1Factory.Function create()
HashFunctionFactory
HashFunction
.HashFunction
protected FloatFVComparison fvDistanceFunction()
fvDistanceFunction
in class FloatHashFunctionFactory