water.bindings.pojos

Class TargetEncoderParametersV3

java.lang.Object

water.bindings.pojos.SchemaV3

water.bindings.pojos.ModelParametersSchemaV3

water.bindings.pojos.TargetEncoderParametersV3

public class TargetEncoderParametersV3
extends ModelParametersSchemaV3

Field Summary

Fields

Modifier and Type	Field and Description
boolean	blending If true, enables blending of posterior probabilities (computed for a given categorical value) with prior probabilities (computed on the entire set).
java.lang.String[][]	columnsToEncode List of categorical columns or groups of categorical columns to encode.
H2otargetencodingTargetEncoderModelDataLeakageHandlingStrategy	dataLeakageHandling Data leakage handling strategy used to generate the encoding.
double	inflectionPoint Inflection point of the sigmoid used to blend probabilities (see `blending` parameter).
boolean	keepOriginalCategoricalColumns If true, the original non-encoded categorical features will remain in the result frame.
double	noise The amount of noise to add to the encoded column.
long	seed Seed used to generate the noise.
double	smoothing Smoothing factor corresponds to the inverse of the slope at the inflection point on the sigmoid used to blend probabilities (see `blending` parameter).

Fields inherited from class water.bindings.pojos.ModelParametersSchemaV3

aucType, categoricalEncoding, checkpoint, customDistributionFunc, customMetricFunc, distribution, exportCheckpointsDir, foldAssignment, foldColumn, gainsliftBins, huberAlpha, ignoreConstCols, ignoredColumns, keepCrossValidationFoldAssignment, keepCrossValidationModels, keepCrossValidationPredictions, maxCategoricalLevels, maxRuntimeSecs, modelId, nfolds, offsetColumn, parallelizeCrossValidation, quantileAlpha, responseColumn, scoreEachIteration, stoppingMetric, stoppingRounds, stoppingTolerance, trainingFrame, tweediePower, validationFrame, weightsColumn

Constructor Summary

Constructors

Constructor and Description
TargetEncoderParametersV3() Public constructor

Method Summary

Modifier and Type	Method and Description
java.lang.String	toString() Return the contents of this object as a JSON String.

Methods inherited from class java.lang.Object

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

Field Detail

columnsToEncode

@SerializedName(value="columns_to_encode")
public java.lang.String[][] columnsToEncode

List of categorical columns or groups of categorical columns to encode. When groups of columns are specified, each group is encoded as a single column (interactions are created internally).

keepOriginalCategoricalColumns

@SerializedName(value="keep_original_categorical_columns")
public boolean keepOriginalCategoricalColumns

If true, the original non-encoded categorical features will remain in the result frame.

blending

public boolean blending

If true, enables blending of posterior probabilities (computed for a given categorical value) with prior probabilities (computed on the entire set). This allows to mitigate the effect of categorical values with small cardinality. The blending effect can be tuned using the `inflection_point` and `smoothing` parameters.

inflectionPoint

@SerializedName(value="inflection_point")
public double inflectionPoint

Inflection point of the sigmoid used to blend probabilities (see `blending` parameter). For a given categorical value, if it appears less that `inflection_point` in a data sample, then the influence of the posterior probability will be smaller than the prior.

smoothing

public double smoothing

Smoothing factor corresponds to the inverse of the slope at the inflection point on the sigmoid used to blend probabilities (see `blending` parameter). If smoothing tends towards 0, then the sigmoid used for blending turns into a Heaviside step function.

dataLeakageHandling

@SerializedName(value="data_leakage_handling")
public H2otargetencodingTargetEncoderModelDataLeakageHandlingStrategy dataLeakageHandling

Data leakage handling strategy used to generate the encoding. Supported options are: 1) "none" (default) - no holdout, using the entire training frame. 2) "leave_one_out" - current row's response value is subtracted from the per-level frequencies pre-calculated on the entire training frame. 3) "k_fold" - encodings for a fold are generated based on out-of-fold data.

noise

public double noise

The amount of noise to add to the encoded column. Use 0 to disable noise, and -1 (=AUTO) to let the algorithm determine a reasonable amount of noise.

seed

public long seed

Seed used to generate the noise. By default, the seed is chosen randomly.

Constructor Detail

TargetEncoderParametersV3

public TargetEncoderParametersV3()

Public constructor

Method Detail

toString

public java.lang.String toString()

Return the contents of this object as a JSON String.

Overrides:

toString in class ModelParametersSchemaV3