Package com.tagtraum.core.ml

Class RandomForest

java.lang.Object

com.tagtraum.core.ml.RandomForest

Direct Known Subclasses:

RandomForestClassifier, RandomForestRegressor

public abstract class RandomForest
extends Object

RandomForest.

Author:

Hendrik Schreiber

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
protected static class	RandomForest.Node	Node.

Field Summary

Fields

Modifier and Type	Field	Description
protected List<RandomForest.Node>	trees

Constructor Summary

Constructors

Constructor	Description
RandomForest(URL model)	Creates a Random Forest based on the provided model file.

Method Summary

Modifier and Type	Method	Description
abstract float	predict(float[] features)
byte[]	toBinary()	Create binary representation of this forest.
String	toDebugString()	Human readable debug string.
String	toString()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

trees

protected final List<RandomForest.Node> trees

Constructor Detail

RandomForest

public RandomForest(URL model)
             throws IOException

Creates a Random Forest based on the provided model file. The file may be a text file, like it is produced by Spark's toDebugString(). Additionally, gzip compression is supported, if the file name end's in .gz, and a proprietary binary format generated by toBinary().

Parameters:

model - model file

Throws:

IOException

Method Detail

predict

public abstract float predict(float[] features)

toDebugString

public String toDebugString()

Human readable debug string.

Returns:

string

toBinary

public byte[] toBinary()

Create binary representation of this forest.

Returns:

binary representation

toString

public String toString()

Overrides:

toString in class Object