hex.singlenoderf

Class CMTask

java.lang.Object

jsr166y.ForkJoinTask<java.lang.Void>

jsr166y.CountedCompleter

water.H2O.H2OCountedCompleter

water.DTask

water.MRTask2<CMTask>

hex.singlenoderf.CMTask

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, java.util.concurrent.Future<java.lang.Void>, jsr166y.ForkJoinPool.ManagedBlocker, Freezable

public class CMTask
extends MRTask2<CMTask>

Confusion Matrix. Incrementally computes a Confusion Matrix for a forest of Trees, vs a given input dataset. The set of Trees can grow over time. Each request from the Confusion compute on any new trees (if any), and report a matrix. Cheap if all trees already computed.

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	CMTask.CMFinal

Field Summary

Fields

Modifier and Type	Field and Description
int	_classcol
double[]	_classWt
long[][][]	_cms
boolean	_computeOOB
Frame	_data
Key	_datakey
long[]	_errorsPerTree
hex.singlenoderf.CMTask.CM[]	_localMatrices
hex.singlenoderf.CMTask.CM	_matrix
SpeeDRFModel	_model
int[]	_modelDataMap
Key	_modelKey
int	_N
int[]	_oobs
float	_sum
int	_treesUsed
VarImp	_varimp

Fields inherited from class water.MRTask2

_appendables, _fr, _fs, _hi, _left, _lo, _nhi, _nleft, _nrite, _nxx, _rite, _t0, _topLocal

Fields inherited from class water.DTask

_cls, _eFromNode, _exception, _fname, _lineNum, _msg, _mth

Constructor Summary

Constructors

Constructor and Description
CMTask(SpeeDRFModel model, int treesToUse, double[] classWt, boolean computeOOB) Confusion matrix

Method Summary

Methods

Modifier and Type	Method and Description
static int	alignEnumDomains(java.lang.String[] modelDomain, java.lang.String[] dataDomain, int[] modelMapping, int[] dataMapping) Merge model and data predictor domain to produce domain for CM.
static float[]	computeVarImpSD(long[][] vote_diffs)
int	dimension() Return number of classes - in fact dimension of CM.
java.lang.String[]	domain() Compute confusion matrix domain based on model and data key.
static java.lang.String[]	domain(int N, Vec modelCol, Vec dataCol, int[] modelEnumMapping, int[] dataEnumMapping)
static java.lang.String[]	domain(Vec modelCol, Vec dataCol)
void	init()
void	map(Chunk[] chks) Override with your map implementation.
void	reduce(CMTask drt) Reduction combines the confusion matrices.

Methods inherited from class water.MRTask2

asyncExec, asyncExec, asyncExec, block, clone, closeLocal, compute2, dfork, dfork, dfork, dfork, dinvoke, doAll, doAll, doAll, doAll, doAll, doAll, exec, exec, exec, getResult, isReleasable, map, map, map, map, map, map, map, map, map, map, map, onCompletion, onExceptionalCompletion, outputFrame, outputFrame, outputFrame, postGlobal, priority, profString, reduce4, setupLocal, vecs

Methods inherited from class water.DTask

copyOver, frozenType, getDException, hasException, logVerbose, newInstance, onAck, onAckAck, read, setException, toDocField, write, writeJSONFields

Methods inherited from class water.H2O.H2OCountedCompleter

compute

Methods inherited from class jsr166y.CountedCompleter

addToPendingCount, compareAndSetPendingCount, complete, exec, getCompleter, getPendingCount, getRawResult, setCompleter, setPendingCount, setRawResult, tryComplete

Methods inherited from class jsr166y.ForkJoinTask

adapt, adapt, adapt, cancel, compareAndSetForkJoinTaskTag, completeExceptionally, fork, get, get, getException, getForkJoinTaskTag, getPool, getQueuedTaskCount, getSurplusQueuedTaskCount, helpQuiesce, inForkJoinPool, invoke, invokeAll, invokeAll, invokeAll, isCancelled, isCompletedAbnormally, isCompletedNormally, isDone, join, peekNextLocalTask, pollNextLocalTask, pollTask, quietlyComplete, quietlyInvoke, quietlyJoin, reinitialize, setForkJoinTaskTag, tryUnfork

Methods inherited from class java.lang.Object

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

Field Detail

_classWt

public double[] _classWt

_computeOOB

public boolean _computeOOB

_treesUsed

public int _treesUsed

_modelKey

public Key _modelKey

_datakey

public Key _datakey

_classcol

public int _classcol

_matrix

public hex.singlenoderf.CMTask.CM _matrix

_sum

public float _sum

_localMatrices

public hex.singlenoderf.CMTask.CM[] _localMatrices

_errorsPerTree

public long[] _errorsPerTree

_model

public SpeeDRFModel _model

_modelDataMap

public int[] _modelDataMap

_data

public Frame _data

_N

public int _N

_cms

public long[][][] _cms

_varimp

public VarImp _varimp

_oobs

public int[] _oobs

Constructor Detail

CMTask

public CMTask(SpeeDRFModel model,
      int treesToUse,
      double[] classWt,
      boolean computeOOB)

Confusion matrix

Parameters:

model - the ensemble used to classify

Method Detail

init

public void init()

map

public void map(Chunk[] chks)

Description copied from class: MRTask2

Override with your map implementation. This overload is given an array of local input Chunks, for Frames with arbitrary column numbers. All map variants are called, but only one is expected to be overridden.

Overrides:

map in class MRTask2<CMTask>

computeVarImpSD

public static float[] computeVarImpSD(long[][] vote_diffs)

reduce

public void reduce(CMTask drt)

Reduction combines the confusion matrices.

Overrides:

reduce in class MRTask2<CMTask>

alignEnumDomains

public static int alignEnumDomains(java.lang.String[] modelDomain,
                   java.lang.String[] dataDomain,
                   int[] modelMapping,
                   int[] dataMapping)

Merge model and data predictor domain to produce domain for CM. The domain is expected to be ordered and containing unique values.

domain

public static java.lang.String[] domain(Vec modelCol,
                        Vec dataCol)

domain

public static java.lang.String[] domain(int N,
                        Vec modelCol,
                        Vec dataCol,
                        int[] modelEnumMapping,
                        int[] dataEnumMapping)

domain

public java.lang.String[] domain()

Compute confusion matrix domain based on model and data key.

dimension

public final int dimension()

Return number of classes - in fact dimension of CM.