RPUtils (deeplearning4j 1.0.0-beta6 API)

java.lang.Object
- org.deeplearning4j.clustering.randomprojection.RPUtils

```
public class RPUtils
extends Object
```
A port of: https://github.com/lyst/rpforest to nd4j

Author:

Constructor Summary

Constructors
Constructor and Description

RPUtils()

Constructors
Constructor and Description
`RPUtils()`

Method Summary

All Methods Static Methods Concrete Methods
Modifier and Type	Method and Description
`static void`	`buildTree(RPTree tree, RPNode from, RPHyperPlanes planes, INDArray X, int maxSize, int depth, String similarityFunction)` Initialize the tree given the input parameters
`static double`	`computeDistance(String function, INDArray x, INDArray y)` Compute the distance between 2 vectors given a function name.
`static double`	`computeDistance(String function, INDArray x, INDArray y, INDArray result)` /** Compute the distance between 2 vectors given a function name.
`static INDArray`	`computeDistanceMulti(String function, INDArray x, INDArray y, INDArray result)` Compute the distance between 2 vectors given a function name.
`static INDArray`	`getAllCandidates(INDArray x, List<RPTree> trees, String similarityFunction)` Get the search candidates as indices given the input and similarity function
`static List<Integer>`	`getCandidates(INDArray x, List<RPTree> roots, String similarityFunction)` Get the search candidates as indices given the input and similarity function
`static <T extends DifferentialFunction> DifferentialFunction`	`getOp(String name, INDArray x, INDArray y, INDArray result)`
`static RPNode`	`query(RPNode from, RPHyperPlanes planes, INDArray x, String similarityFunction)` Query the tree starting from the given node using the given hyper plane and similarity function
`static INDArray`	`queryAll(INDArray toQuery, INDArray X, List<RPTree> trees, int n, String similarityFunction)` Query all trees using the given input and data
`static List<Pair<Double,Integer>>`	`queryAllWithDistances(INDArray toQuery, INDArray X, List<RPTree> trees, int n, String similarityFunction)` Query all trees using the given input and data
`static void`	`scanForLeaves(List<RPNode> nodes, RPNode current)` Scan for leaves accumulating the nodes in the passed in list
`static void`	`scanForLeaves(List<RPNode> nodes, RPTree scan)` Scan for leaves accumulating the nodes in the passed in list
`static void`	`slimNode(RPNode node)` Prune indices from the given node when it's a leaf
`static List<Pair<Double,Integer>>`	`sortCandidates(INDArray x, INDArray X, List<Integer> candidates, String similarityFunction)` Get the sorted distances given the query vector, input data, given the list of possible search candidates

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail
- RPUtils
```
public RPUtils()
```

Method Detail

getOp

public static <T extends DifferentialFunction> DifferentialFunction getOp(String name,
                                                                          INDArray x,
                                                                          INDArray y,
                                                                          INDArray result)

queryAllWithDistances

public static List<Pair<Double,Integer>> queryAllWithDistances(INDArray toQuery,
                                                               INDArray X,
                                                               List<RPTree> trees,
                                                               int n,
                                                               String similarityFunction)

Query all trees using the given input and data

Parameters:: toQuery - the query vector; X - the input data to query; trees - the trees to query; n - the number of results to search for; similarityFunction - the similarity function to use
Returns:: the indices (in order) in the ndarray

queryAll

public static INDArray queryAll(INDArray toQuery,
                                INDArray X,
                                List<RPTree> trees,
                                int n,
                                String similarityFunction)

Query all trees using the given input and data

Parameters:: toQuery - the query vector; X - the input data to query; trees - the trees to query; n - the number of results to search for; similarityFunction - the similarity function to use
Returns:: the indices (in order) in the ndarray

sortCandidates

public static List<Pair<Double,Integer>> sortCandidates(INDArray x,
                                                        INDArray X,
                                                        List<Integer> candidates,
                                                        String similarityFunction)

Get the sorted distances given the query vector, input data, given the list of possible search candidates

Parameters:: x - the query vector; X - the input data to use; candidates - the possible search candidates; similarityFunction - the similarity function to use
Returns:: the sorted distances

getAllCandidates

public static INDArray getAllCandidates(INDArray x,
                                        List<RPTree> trees,
                                        String similarityFunction)

Get the search candidates as indices given the input and similarity function

Parameters:: x - the input data to search with; trees - the trees to search; similarityFunction - the function to use for similarity
Returns:: the list of indices as the search results

getCandidates

public static List<Integer> getCandidates(INDArray x,
                                          List<RPTree> roots,
                                          String similarityFunction)

Get the search candidates as indices given the input and similarity function

Parameters:: x - the input data to search with; roots - the trees to search; similarityFunction - the function to use for similarity
Returns:: the list of indices as the search results

query
```
public static RPNode query(RPNode from,
                           RPHyperPlanes planes,
                           INDArray x,
                           String similarityFunction)
```
Query the tree starting from the given node using the given hyper plane and similarity function

Parameters:

from - the node to start from

planes - the hyper plane to query

x - the input data

similarityFunction - the similarity function to use

Returns:

the leaf node representing the given query from a search in the tree

computeDistanceMulti
```
public static INDArray computeDistanceMulti(String function,
                                            INDArray x,
                                            INDArray y,
                                            INDArray result)
```
Compute the distance between 2 vectors given a function name. Valid function names: euclidean: euclidean distance cosinedistance: cosine distance cosine similarity: cosine similarity manhattan: manhattan distance jaccard: jaccard distance hamming: hamming distance

Parameters:

function - the function to use (default euclidean distance)

x - the first vector

y - the second vector

Returns:

the distance between the 2 vectors given the inputs

computeDistance
```
public static double computeDistance(String function,
                                     INDArray x,
                                     INDArray y,
                                     INDArray result)
```
/** Compute the distance between 2 vectors given a function name. Valid function names: euclidean: euclidean distance cosinedistance: cosine distance cosine similarity: cosine similarity manhattan: manhattan distance jaccard: jaccard distance hamming: hamming distance

Parameters:

function - the function to use (default euclidean distance)

x - the first vector

y - the second vector

Returns:

the distance between the 2 vectors given the inputs

computeDistance
```
public static double computeDistance(String function,
                                     INDArray x,
                                     INDArray y)
```
Compute the distance between 2 vectors given a function name. Valid function names: euclidean: euclidean distance cosinedistance: cosine distance cosine similarity: cosine similarity manhattan: manhattan distance jaccard: jaccard distance hamming: hamming distance

Parameters:

function - the function to use (default euclidean distance)

x - the first vector

y - the second vector

Returns:

the distance between the 2 vectors given the inputs

buildTree

public static void buildTree(RPTree tree,
                             RPNode from,
                             RPHyperPlanes planes,
                             INDArray X,
                             int maxSize,
                             int depth,
                             String similarityFunction)

Initialize the tree given the input parameters

Parameters:: tree - the tree to initialize; from - the starting node; planes - the hyper planes to use (vector space for similarity); X - the input data; maxSize - the max number of indices on a given leaf node; depth - the current depth of the tree; similarityFunction - the similarity function to use

scanForLeaves

public static void scanForLeaves(List<RPNode> nodes,
                                 RPTree scan)

Scan for leaves accumulating the nodes in the passed in list

Parameters:: nodes - the nodes so far; scan - the tree to scan

scanForLeaves

public static void scanForLeaves(List<RPNode> nodes,
                                 RPNode current)

Scan for leaves accumulating the nodes in the passed in list

Parameters:: nodes - the nodes so far

slimNode
```
public static void slimNode(RPNode node)
```
Prune indices from the given node when it's a leaf

Parameters:

node - the node to prune

Class RPUtils

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Constructor Detail

RPUtils

Method Detail

getOp

queryAllWithDistances

queryAll

sortCandidates

getAllCandidates

getCandidates

query

computeDistanceMulti

computeDistance

computeDistance

buildTree

scanForLeaves

scanForLeaves

slimNode