EasyPredictModelWrapper (h2o-genmodel 3.46.0 API)

java.lang.Object
- hex.genmodel.easy.EasyPredictModelWrapper

All Implemented Interfaces:

java.io.Serializable
```
public class EasyPredictModelWrapper
extends java.lang.Object
implements java.io.Serializable
```
An easy-to-use prediction wrapper for generated models. Instantiate as follows. The following two are equivalent. EasyPredictModelWrapper model = new EasyPredictModelWrapper(rawModel); EasyPredictModelWrapper model = new EasyPredictModelWrapper( new EasyPredictModelWrapper.Config() .setModel(rawModel) .setConvertUnknownCategoricalLevelsToNa(false)); Note that for any given model, you must use the exact one correct predict method below based on the model category. By default, unknown categorical levels result in a thrown PredictUnknownCategoricalLevelException. The API was designed with this default to make the simplest possible setup inform the user if there are concerns with the data quality. An alternate behavior is to automatically convert unknown categorical levels to N/A. To do this, use setConvertUnknownCategoricalLevelsToNa(true) instead. Detection of unknown categoricals may be observed by registering an implementation of EasyPredictModelWrapper.ErrorConsumer in the process of EasyPredictModelWrapper.Config creation. Advanced scoring features are disabled by default for performance reasons. Configuration flags allow the user to output also - leaf node assignment, - GLRM reconstructed matrix, - staged probabilities, - prediction contributions (SHAP values). Deprecation note: Total number of unknown categorical variables is newly accessible by registering CountingErrorConsumer.
See the top-of-tree master version of this file here on github.

See Also:

Serialized Form

Nested Class Summary

Nested Classes
Modifier and Type	Class and Description
`static class`	`EasyPredictModelWrapper.Config` Configuration builder for instantiating a Wrapper.
`static class`	`EasyPredictModelWrapper.ErrorConsumer` Observer interface with methods corresponding to errors during the prediction.

Field Summary

Fields
Modifier and Type Field and Description

GenModel m

Fields
Modifier and Type	Field and Description
`GenModel`	`m`

Constructor Summary

Constructors
Constructor and Description
`EasyPredictModelWrapper(EasyPredictModelWrapper.Config config)` Create a wrapper for a generated model.
`EasyPredictModelWrapper(GenModel model)` Create a wrapper for a generated model.

Method Summary

All Methods Instance Methods Concrete Methods Deprecated Methods
Modifier and Type	Method and Description
`protected double[]`	`fillRawData(RowData data, double[] rawData)`
`java.lang.String[]`	`getContributionNames()` Returns names of contributions for prediction results with constributions enabled.
`boolean`	`getEnableContributions()`
`boolean`	`getEnableGLRMReconstruct()`
`boolean`	`getEnableLeafAssignment()`
`boolean`	`getEnableStagedProbabilities()`
`java.lang.String`	`getHeader()` Some autoencoder thing, I'm not sure what this does.
`GenModel`	`getModel()`
`ModelCategory`	`getModelCategory()` Get the category (type) of model.
`java.lang.String[]`	`getResponseDomainValues()` Get the array of levels for the response column.
`java.lang.String[]`	`leafNodeAssignment(RowData data)`
`SharedTreeMojoModel.LeafNodeAssignments`	`leafNodeAssignmentExtended(RowData data)`
`protected double[]`	`preamble(ModelCategory c, RowData data)`
`protected double[]`	`preamble(ModelCategory c, RowData data, double offset)`
`AbstractPrediction`	`predict(RowData data)` See `predict(RowData, ModelCategory)`
`protected double[]`	`predict(RowData data, double offset, double[] preds)`
`AbstractPrediction`	`predict(RowData data, ModelCategory mc)` Make a prediction on a new data point.
`AnomalyDetectionPrediction`	`predictAnomalyDetection(RowData data)` Make a prediction on a new data point using a Anomaly Detection model.
`AutoEncoderModelPrediction`	`predictAutoEncoder(RowData data)` Make a prediction on a new data point using an AutoEncoder model.
`BinomialModelPrediction`	`predictBinomial(RowData data)` Make a prediction on a new data point using a Binomial model.
`BinomialModelPrediction`	`predictBinomial(RowData data, double offset)` Make a prediction on a new data point using a Binomial model.
`ClusteringModelPrediction`	`predictClustering(RowData data)` Make a prediction on a new data point using a Clustering model.
`float[]`	`predictContributions(RowData data)`
`FeatureContribution[]`	`predictContributions(RowData data, int topN, int bottomN, boolean compareAbs)` Calculate and sort shapley values.
`CoxPHModelPrediction`	`predictCoxPH(RowData data)`
`CoxPHModelPrediction`	`predictCoxPH(RowData data, double offset)`
`DimReductionModelPrediction`	`predictDimReduction(RowData data)` Make a prediction on a new data point using a Dimension Reduction model (PCA, GLRM)
`KLimeModelPrediction`	`predictKLime(RowData data)`
`MultinomialModelPrediction`	`predictMultinomial(RowData data)` Make a prediction on a new data point using a Multinomial model.
`MultinomialModelPrediction`	`predictMultinomial(RowData data, double offset)` Make a prediction on a new data point using a Multinomial model.
`OrdinalModelPrediction`	`predictOrdinal(RowData data)` Make a prediction on a new data point using a Ordinal model.
`OrdinalModelPrediction`	`predictOrdinal(RowData data, double offset)` Make a prediction on a new data point using a Ordinal model.
`double[]`	`predictRaw(RowData data, double offset)` Make a prediction on a new data point.
`RegressionModelPrediction`	`predictRegression(RowData data)` Make a prediction on a new data point using a Regression model.
`RegressionModelPrediction`	`predictRegression(RowData data, double offset)` Make a prediction on a new data point using a Regression model.
`TargetEncoderPrediction`	`predictTargetEncoding(RowData data)` Perform target encoding based on TargetEncoderMojoModel
`UpliftBinomialModelPrediction`	`predictUpliftBinomial(RowData data)` Make a prediction on a new data point using Uplift Binomial model.
`Word2VecPrediction`	`predictWord2Vec(RowData data)` Lookup word embeddings for a given word (or set of words).
`float[]`	`predictWord2Vec(java.lang.String[] sentence)` Calculate an aggregated word-embedding for a given input sentence (sequence of words).
`SortedClassProbability[]`	`sortByDescendingClassProbability(BinomialModelPrediction p)` A helper function to return an array of binomial class probabilities for a prediction in sorted order.
`TargetEncoderPrediction`	`transformWithTargetEncoding(RowData data)` Deprecated. Use `predictTargetEncoding(RowData)` instead.
`KeyValue[]`	`varimp()` See `varimp(int)` return descending sorted by relative importance array of all variables in the model
`KeyValue[]`	`varimp(int n)` See `VariableImportances.topN(int)`

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - m
```
public final GenModel m
```
- Constructor Detail
  - EasyPredictModelWrapper
```
public EasyPredictModelWrapper(EasyPredictModelWrapper.Config config)
```
    Create a wrapper for a generated model.
    
    Parameters:
    
    config - The wrapper configuration
  - EasyPredictModelWrapper
```
public EasyPredictModelWrapper(GenModel model)
```
    Create a wrapper for a generated model.
    
    Parameters:
    
    model - The generated model
- Method Detail
  - getEnableLeafAssignment
```
public boolean getEnableLeafAssignment()
```
  - getEnableGLRMReconstruct
```
public boolean getEnableGLRMReconstruct()
```
  - getEnableStagedProbabilities
```
public boolean getEnableStagedProbabilities()
```
  - getEnableContributions
```
public boolean getEnableContributions()
```
  - predict
```
public AbstractPrediction predict(RowData data,
                                  ModelCategory mc)
                           throws PredictException
```
    Make a prediction on a new data point. The type of prediction returned depends on the model type. The caller needs to know what type of prediction to expect. This call is convenient for generically automating model deployment. For specific applications (where the kind of model is known and doesn't change), it is recommended to call specific prediction calls like predictBinomial() directly.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictRaw
```
public double[] predictRaw(RowData data,
                           double offset)
                    throws PredictException
```
    Make a prediction on a new data point. This method has the same input as predict. The only difference is that it returns and array instead of a prediction object. The meaning of the returned values can be decoded by calling getOutputNames and if any returned values are categorical - method getOutputDomain can be used to find mapping of indexes to categorical values for the particular column.
    
    Parameters:
    
    data - A new data point. Column names are case-sensitive.
    
    offset - Value of offset (use 0 if the model was trained without offset).
    
    Returns:
    
    An array representing a prediction.
    
    Throws:
    
    PredictException - if prediction cannot be made (eg.: input is invalid)
  - predict
```
public AbstractPrediction predict(RowData data)
                           throws PredictException
```
    See predict(RowData, ModelCategory)
    
    Throws:
    
    PredictException
  - getContributionNames
```
public java.lang.String[] getContributionNames()
```
    Returns names of contributions for prediction results with constributions enabled.
    
    Returns:
    
    array of contribution names (array has same lenght as the actual contributions, last is BiasTerm)
  - predictAutoEncoder
```
public AutoEncoderModelPrediction predictAutoEncoder(RowData data)
                                              throws PredictException
```
    Make a prediction on a new data point using an AutoEncoder model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictDimReduction
```
public DimReductionModelPrediction predictDimReduction(RowData data)
                                                throws PredictException
```
    Make a prediction on a new data point using a Dimension Reduction model (PCA, GLRM)
    
    Parameters:
    
    data - A new data point. Unknown column name is treated as a NaN. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictWord2Vec
```
public float[] predictWord2Vec(java.lang.String[] sentence)
                        throws PredictException
```
    Calculate an aggregated word-embedding for a given input sentence (sequence of words).
    
    Parameters:
    
    sentence - array of word forming a sentence
    
    Returns:
    
    word-embedding for the given sentence calculated by averaging the embeddings of the input words
    
    Throws:
    
    PredictException - if model is not a WordEmbedding model
  - predictWord2Vec
```
public Word2VecPrediction predictWord2Vec(RowData data)
                                   throws PredictException
```
    Lookup word embeddings for a given word (or set of words). The result is a dictionary of words mapped to their respective embeddings.
    
    Parameters:
    
    data - RawData structure, every key with a String value will be translated to an embedding, note: keys only purpose is to link the output embedding to the input word.
    
    Returns:
    
    The prediction
    
    Throws:
    
    PredictException - if model is not a WordEmbedding model
  - predictAnomalyDetection
```
public AnomalyDetectionPrediction predictAnomalyDetection(RowData data)
                                                   throws PredictException
```
    Make a prediction on a new data point using a Anomaly Detection model.
    
    Parameters:
    
    data - A new data point. Unknown column name is treated as a NaN. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictBinomial
```
public BinomialModelPrediction predictBinomial(RowData data)
                                        throws PredictException
```
    Make a prediction on a new data point using a Binomial model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictBinomial
```
public BinomialModelPrediction predictBinomial(RowData data,
                                               double offset)
                                        throws PredictException
```
    Make a prediction on a new data point using a Binomial model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    offset - An offset for the prediction.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictUpliftBinomial
```
public UpliftBinomialModelPrediction predictUpliftBinomial(RowData data)
                                                    throws PredictException
```
    Make a prediction on a new data point using Uplift Binomial model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - transformWithTargetEncoding
```
@Deprecated
public TargetEncoderPrediction transformWithTargetEncoding(RowData data)
                                                                throws PredictException
```
    Deprecated. Use predictTargetEncoding(RowData) instead.
    
    Throws:
    
    PredictException
  - predictTargetEncoding
```
public TargetEncoderPrediction predictTargetEncoding(RowData data)
                                              throws PredictException
```
    Perform target encoding based on TargetEncoderMojoModel
    
    Parameters:
    
    data - RowData structure with data for which we want to produce transformations. Unknown column name is treated as a NaN. Column names are case sensitive.
    
    Returns:
    
    TargetEncoderPrediction with transformations ordered in accordance with corresponding categorical columns' indices in training data
    
    Throws:
    
    PredictException
  - leafNodeAssignment
```
public java.lang.String[] leafNodeAssignment(RowData data)
                                      throws PredictException
```
    Throws:
    
    PredictException
  - leafNodeAssignmentExtended
```
public SharedTreeMojoModel.LeafNodeAssignments leafNodeAssignmentExtended(RowData data)
                                                                   throws PredictException
```
    Throws:
    
    PredictException
  - predictMultinomial
```
public MultinomialModelPrediction predictMultinomial(RowData data)
                                              throws PredictException
```
    Make a prediction on a new data point using a Multinomial model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictMultinomial
```
public MultinomialModelPrediction predictMultinomial(RowData data,
                                                     double offset)
                                              throws PredictException
```
    Make a prediction on a new data point using a Multinomial model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    offset - Prediction offset
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictOrdinal
```
public OrdinalModelPrediction predictOrdinal(RowData data)
                                      throws PredictException
```
    Make a prediction on a new data point using a Ordinal model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictOrdinal
```
public OrdinalModelPrediction predictOrdinal(RowData data,
                                             double offset)
                                      throws PredictException
```
    Make a prediction on a new data point using a Ordinal model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    offset - Prediction offset
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - sortByDescendingClassProbability
```
public SortedClassProbability[] sortByDescendingClassProbability(BinomialModelPrediction p)
```
    A helper function to return an array of binomial class probabilities for a prediction in sorted order. The returned array has the most probable class in position 0.
    
    Parameters:
    
    p - The prediction.
    
    Returns:
    
    An array with sorted class probabilities.
  - predictClustering
```
public ClusteringModelPrediction predictClustering(RowData data)
                                            throws PredictException
```
    Make a prediction on a new data point using a Clustering model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictRegression
```
public RegressionModelPrediction predictRegression(RowData data)
                                            throws PredictException
```
    Make a prediction on a new data point using a Regression model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictRegression
```
public RegressionModelPrediction predictRegression(RowData data,
                                                   double offset)
                                            throws PredictException
```
    Make a prediction on a new data point using a Regression model.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    offset - Prediction offset
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictKLime
```
public KLimeModelPrediction predictKLime(RowData data)
                                  throws PredictException
```
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    Returns:
    
    The prediction.
    
    Throws:
    
    PredictException
  - predictCoxPH
```
public CoxPHModelPrediction predictCoxPH(RowData data,
                                         double offset)
                                  throws PredictException
```
    Throws:
    
    PredictException
  - predictCoxPH
```
public CoxPHModelPrediction predictCoxPH(RowData data)
                                  throws PredictException
```
    Throws:
    
    PredictException
  - predictContributions
```
public float[] predictContributions(RowData data)
                             throws PredictException
```
    Throws:
    
    PredictException
  - predictContributions
```
public FeatureContribution[] predictContributions(RowData data,
                                                  int topN,
                                                  int bottomN,
                                                  boolean compareAbs)
                                           throws PredictException
```
    Calculate and sort shapley values.
    
    Parameters:
    
    data - A new data point. Unknown or missing column name is treated as a NaN or ignored. Column names are case sensitive.
    
    topN - Return only #topN highest contributions + bias. If topN<0 then sort all SHAP values in descending order If topN<0 && bottomN<0 then sort all SHAP values in descending order
    
    bottomN - Return only #bottomN lowest contributions + bias If topN and bottomN are defined together then return array of #topN + #bottomN + bias If bottomN<0 then sort all SHAP values in ascending order If topN<0 && bottomN<0 then sort all SHAP values in descending order
    
    compareAbs - True to compare absolute values of contributions
    
    Returns:
    
    Sorted FeatureContribution array of contributions of size #topN + #bottomN + bias If topN < 0 || bottomN < 0 then all descending/ascending sorted contributions is returned.
    
    Throws:
    
    PredictException - When #data cannot be properly translate to raw data.
  - varimp
```
public KeyValue[] varimp()
```
    See varimp(int) return descending sorted by relative importance array of all variables in the model
  - varimp
```
public KeyValue[] varimp(int n)
```
    See VariableImportances.topN(int)
  - getModel
```
public GenModel getModel()
```
  - getModelCategory
```
public ModelCategory getModelCategory()
```
    Get the category (type) of model.
    
    Returns:
    
    The category.
  - getResponseDomainValues
```
public java.lang.String[] getResponseDomainValues()
```
    Get the array of levels for the response column. "Domain" just means list of level names for a categorical (aka factor, enum) column. If the response column is numerical and not categorical, this will return null.
    
    Returns:
    
    The array.
  - getHeader
```
public java.lang.String getHeader()
```
    Some autoencoder thing, I'm not sure what this does.
    
    Returns:
    
    CSV header for autoencoder.
  - preamble
```
protected double[] preamble(ModelCategory c,
                            RowData data)
                     throws PredictException
```
    Throws:
    
    PredictException
  - preamble
```
protected double[] preamble(ModelCategory c,
                            RowData data,
                            double offset)
                     throws PredictException
```
    Throws:
    
    PredictException
  - fillRawData
```
protected double[] fillRawData(RowData data,
                               double[] rawData)
                        throws PredictException
```
    Throws:
    
    PredictException
  - predict
```
protected double[] predict(RowData data,
                           double offset,
                           double[] preds)
                    throws PredictException
```
    Throws:
    
    PredictException

Class EasyPredictModelWrapper

Nested Class Summary

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

m

Constructor Detail

EasyPredictModelWrapper

EasyPredictModelWrapper

Method Detail

getEnableLeafAssignment

getEnableGLRMReconstruct

getEnableStagedProbabilities

getEnableContributions

predict

predictRaw

predict

getContributionNames

predictAutoEncoder

predictDimReduction

predictWord2Vec

predictWord2Vec

predictAnomalyDetection

predictBinomial

predictBinomial

predictUpliftBinomial

transformWithTargetEncoding

predictTargetEncoding

leafNodeAssignment

leafNodeAssignmentExtended

predictMultinomial

predictMultinomial

predictOrdinal

predictOrdinal

sortByDescendingClassProbability

predictClustering

predictRegression

predictRegression

predictKLime

predictCoxPH

predictCoxPH

predictContributions

predictContributions

varimp

varimp

getModel

getModelCategory

getResponseDomainValues

getHeader

preamble

preamble

fillRawData

predict