hex.tree

Class SharedTreeModel.SharedTreeOutput

java.lang.Object

water.Iced

hex.Model.Output

hex.SupervisedModel.SupervisedOutput

hex.tree.SharedTreeModel.SharedTreeOutput

All Implemented Interfaces:

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

Direct Known Subclasses:

DRFModel.DRFOutput, GBMModel.GBMOutput

Enclosing class:

SharedTreeModel<M extends SharedTreeModel<M,P,O>,P extends SharedTreeModel.SharedTreeParameters,O extends SharedTreeModel.SharedTreeOutput>

public abstract static class SharedTreeModel.SharedTreeOutput
extends hex.SupervisedModel.SupervisedOutput

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
double	_init_f InitF value (for zero trees) f0 = mean(yi) for gaussian f0 = log(yi/1-yi) for bernoulli For GBM bernoulli, the initial prediction for 0 trees is p = 1/(1+exp(-f0)) From this, the mse for 0 trees can be computed as follows: mean((yi-p)^2) This is what is stored in _mse_train[0]
double[]	_mse_train Train and validation errors per-tree (scored).
double[]	_mse_valid
int	_ntrees Number of trees actually in the model (as opposed to requested)
long[]	_training_time_ms Training time
water.Key<CompressedTree>[][]	_treeKeys Trees get big, so store each one seperately in the DKV.
TreeStats	_treeStats More indepth tree stats
water.util.TwoDimTable	_variable_importances Variable importances computed during training

Fields inherited from class hex.SupervisedModel.SupervisedOutput

_distribution, _modelClassDist, _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
SharedTreeModel.SharedTreeOutput(SharedTree b, double mse_train, double mse_valid)

Method Summary

Methods

Modifier and Type	Method and Description
void	addKTrees(DTree[] trees)
CompressedTree	ctree(int tnum, int knum)
java.lang.String	toStringTree(int tnum, int knum)

Methods inherited from class hex.SupervisedModel.SupervisedOutput

getModelCategory, isClassifier, isSupervised, nclasses, nfeatures

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

_init_f

public double _init_f

InitF value (for zero trees) f0 = mean(yi) for gaussian f0 = log(yi/1-yi) for bernoulli For GBM bernoulli, the initial prediction for 0 trees is p = 1/(1+exp(-f0)) From this, the mse for 0 trees can be computed as follows: mean((yi-p)^2) This is what is stored in _mse_train[0]

_ntrees

public int _ntrees

Number of trees actually in the model (as opposed to requested)

_treeStats

public final TreeStats _treeStats

More indepth tree stats

_treeKeys

public water.Key<CompressedTree>[][] _treeKeys

Trees get big, so store each one seperately in the DKV.

_mse_train

public double[] _mse_train

Train and validation errors per-tree (scored). Zero index is the no-tree error, guessing only the class distribution. Not all trees are scored, NaN represents trees not scored.

_mse_valid

public double[] _mse_valid

_training_time_ms

public long[] _training_time_ms

Training time

_variable_importances

public water.util.TwoDimTable _variable_importances

Variable importances computed during training

Constructor Detail

SharedTreeModel.SharedTreeOutput

public SharedTreeModel.SharedTreeOutput(SharedTree b,
                                double mse_train,
                                double mse_valid)

Method Detail

addKTrees

public void addKTrees(DTree[] trees)

ctree

public CompressedTree ctree(int tnum,
                   int knum)

toStringTree

public java.lang.String toStringTree(int tnum,
                            int knum)