hex

Class SupervisedModel.SupervisedOutput

java.lang.Object

water.Iced

hex.Model.Output

hex.SupervisedModel.SupervisedOutput

All Implemented Interfaces:

java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, Freezable

Enclosing class:

SupervisedModel<M extends SupervisedModel<M,P,O>,P extends SupervisedModel.SupervisedParameters,O extends SupervisedModel.SupervisedOutput>

public abstract static class SupervisedModel.SupervisedOutput
extends Model.Output

Output from all Supervised Models, includes class distribution

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
long[]	_distribution
double[]	_modelClassDist
double[]	_priorClassDist

Fields inherited from class hex.Model.Output

_domains, _model_metrics, _model_summary, _names, _scoring_history, _state, _training_metrics, _validation_metrics

Constructor Summary

Constructors

Constructor and Description
SupervisedModel.SupervisedOutput()
SupervisedModel.SupervisedOutput(SupervisedModelBuilder b) Any final prep-work just before model-building starts, but after the user has clicked "go".

Method Summary

Methods

Modifier and Type	Method and Description
hex.ModelCategory	getModelCategory()
boolean	isClassifier() Is this model a classification model? (v.
boolean	isSupervised()
int	nclasses()
int	nfeatures() Returns number of input features (OK for most unsupervised methods, need to override for supervised!)

Methods inherited from class hex.Model.Output

addModelMetrics, classNames, responseName, toString

Methods inherited from class water.Iced

clone, frozenType, read_impl, read, readExternal, readJSON_impl, readJSON, toJsonString, write_impl, write, writeExternal, writeHTML_impl, writeHTML, writeJSON_impl, writeJSON

Methods inherited from class java.lang.Object

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

Field Detail

_distribution

public long[] _distribution

_priorClassDist

public double[] _priorClassDist

_modelClassDist

public double[] _modelClassDist

Constructor Detail

SupervisedModel.SupervisedOutput

public SupervisedModel.SupervisedOutput()

SupervisedModel.SupervisedOutput

public SupervisedModel.SupervisedOutput(SupervisedModelBuilder b)

Any final prep-work just before model-building starts, but after the user has clicked "go". E.g., converting a response column to an enum touches the entire column (can be expensive), makes a parallel vec (Key/Data leak management issues), and might throw IAE if there are too many classes.

Method Detail

nfeatures

public int nfeatures()

Description copied from class: Model.Output

Returns number of input features (OK for most unsupervised methods, need to override for supervised!)

Overrides:

nfeatures in class Model.Output

Returns:

Returns number of input features

isSupervised

public boolean isSupervised()

Overrides:

isSupervised in class Model.Output

nclasses

public int nclasses()

Overrides:

nclasses in class Model.Output

Returns:

number of classes; illegal to call before setting distribution

isClassifier

public boolean isClassifier()

Description copied from class: Model.Output

Is this model a classification model? (v. a regression or clustering model)

Overrides:

isClassifier in class Model.Output

getModelCategory

public hex.ModelCategory getModelCategory()

Overrides:

getModelCategory in class Model.Output