org.apache.spark.ml.tuning

Class TrainValidationSplit

Object

org.apache.spark.ml.PipelineStage

org.apache.spark.ml.Estimator<TrainValidationSplitModel>

org.apache.spark.ml.tuning.TrainValidationSplit

All Implemented Interfaces:

java.io.Serializable, Logging, Params, HasCollectSubModels, Identifiable, MLWritable

public class TrainValidationSplit
extends Estimator<TrainValidationSplitModel>
implements HasCollectSubModels, MLWritable, Logging

Validation for hyper-parameter tuning. Randomly splits the input dataset into train and validation sets, and uses evaluation metric on the validation set to select the best model. Similar to CrossValidator, but only splits the set once.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
TrainValidationSplit()
TrainValidationSplit(String uid)

Method Summary

Modifier and Type	Method and Description
static Params	clear(Param<?> param)
static BooleanParam	collectSubModels()
TrainValidationSplit	copy(ParamMap extra) Creates a copy of this instance with the same UID and some extra params.
static Param<Estimator<?>>	estimator()
Param<Estimator<?>>	estimator() param for the estimator to be validated
static Param<ParamMap[]>	estimatorParamMaps()
Param<ParamMap[]>	estimatorParamMaps() param for estimator param maps
static Param<Evaluator>	evaluator()
Param<Evaluator>	evaluator() param for the evaluator used to select hyper-parameters that maximize the validated metric
static String	explainParam(Param<?> param)
static String	explainParams()
static ParamMap	extractParamMap()
static ParamMap	extractParamMap(ParamMap extra)
TrainValidationSplitModel	fit(Dataset<?> dataset) Fits a model to the input data.
static <T> scala.Option<T>	get(Param<T> param)
static boolean	getCollectSubModels()
static <T> scala.Option<T>	getDefault(Param<T> param)
static Estimator<?>	getEstimator()
Estimator<?>	getEstimator()
static ParamMap[]	getEstimatorParamMaps()
ParamMap[]	getEstimatorParamMaps()
static Evaluator	getEvaluator()
Evaluator	getEvaluator()
static <T> T	getOrDefault(Param<T> param)
static int	getParallelism()
static Param<Object>	getParam(String paramName)
static long	getSeed()
static double	getTrainRatio()
double	getTrainRatio()
static <T> boolean	hasDefault(Param<T> param)
static boolean	hasParam(String paramName)
static boolean	isDefined(Param<?> param)
static boolean	isSet(Param<?> param)
static TrainValidationSplit	load(String path)
void	logTuningParams(org.apache.spark.ml.util.Instrumentation instrumentation) Instrumentation logging for tuning params including the inner estimator and evaluator info.
static IntParam	parallelism()
static Param<?>[]	params()
static MLReader<TrainValidationSplit>	read()
static void	save(String path)
static LongParam	seed()
static <T> Params	set(Param<T> param, T value)
TrainValidationSplit	setCollectSubModels(boolean value) Whether to collect submodels when fitting.
TrainValidationSplit	setEstimator(Estimator<?> value)
TrainValidationSplit	setEstimatorParamMaps(ParamMap[] value)
TrainValidationSplit	setEvaluator(Evaluator value)
TrainValidationSplit	setParallelism(int value) Set the maximum level of parallelism to evaluate models in parallel.
TrainValidationSplit	setSeed(long value)
TrainValidationSplit	setTrainRatio(double value)
static String	toString()
static DoubleParam	trainRatio()
DoubleParam	trainRatio() Param for ratio between train and validation data.
StructType	transformSchema(StructType schema) :: DeveloperApi ::
StructType	transformSchemaImpl(StructType schema)
String	uid() An immutable unique ID for the object and its derivatives.
MLWriter	write() Returns an MLWriter instance for this ML instance.

Methods inherited from class org.apache.spark.ml.Estimator

fit, fit, fit, fit

Methods inherited from class Object

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

Methods inherited from interface org.apache.spark.ml.param.shared.HasSeed

getSeed, seed

Methods inherited from interface org.apache.spark.ml.param.Params

clear, copyValues, defaultCopy, defaultParamMap, explainParam, explainParams, extractParamMap, extractParamMap, get, getDefault, getOrDefault, getParam, hasDefault, hasParam, isDefined, isSet, paramMap, params, set, set, set, setDefault, setDefault, shouldOwn

Methods inherited from interface org.apache.spark.ml.util.Identifiable

toString

Methods inherited from interface org.apache.spark.ml.param.shared.HasCollectSubModels

collectSubModels, getCollectSubModels

Methods inherited from interface org.apache.spark.ml.util.MLWritable

save

Methods inherited from interface org.apache.spark.internal.Logging

initializeLogging, initializeLogIfNecessary, initializeLogIfNecessary, isTraceEnabled, log_, log, logDebug, logDebug, logError, logError, logInfo, logInfo, logName, logTrace, logTrace, logWarning, logWarning

Constructor Detail

TrainValidationSplit

public TrainValidationSplit(String uid)

TrainValidationSplit

public TrainValidationSplit()

Method Detail

read

public static MLReader<TrainValidationSplit> read()

load

public static TrainValidationSplit load(String path)

toString

public static String toString()

params

public static Param<?>[] params()

explainParam

public static String explainParam(Param<?> param)

explainParams

public static String explainParams()

isSet

public static final boolean isSet(Param<?> param)

isDefined

public static final boolean isDefined(Param<?> param)

hasParam

public static boolean hasParam(String paramName)

getParam

public static Param<Object> getParam(String paramName)

set

public static final <T> Params set(Param<T> param,
                                   T value)

get

public static final <T> scala.Option<T> get(Param<T> param)

clear

public static final Params clear(Param<?> param)

getOrDefault

public static final <T> T getOrDefault(Param<T> param)

getDefault

public static final <T> scala.Option<T> getDefault(Param<T> param)

hasDefault

public static final <T> boolean hasDefault(Param<T> param)

extractParamMap

public static final ParamMap extractParamMap(ParamMap extra)

extractParamMap

public static final ParamMap extractParamMap()

seed

public static final LongParam seed()

getSeed

public static final long getSeed()

estimator

public static Param<Estimator<?>> estimator()

getEstimator

public static Estimator<?> getEstimator()

estimatorParamMaps

public static Param<ParamMap[]> estimatorParamMaps()

getEstimatorParamMaps

public static ParamMap[] getEstimatorParamMaps()

evaluator

public static Param<Evaluator> evaluator()

getEvaluator

public static Evaluator getEvaluator()

trainRatio

public static DoubleParam trainRatio()

getTrainRatio

public static double getTrainRatio()

parallelism

public static IntParam parallelism()

getParallelism

public static int getParallelism()

collectSubModels

public static final BooleanParam collectSubModels()

getCollectSubModels

public static final boolean getCollectSubModels()

save

public static void save(String path)
                 throws java.io.IOException

Throws:

java.io.IOException

uid

public String uid()

Description copied from interface: Identifiable

An immutable unique ID for the object and its derivatives.

Specified by:

uid in interface Identifiable

Returns:

(undocumented)

setEstimator

public TrainValidationSplit setEstimator(Estimator<?> value)

setEstimatorParamMaps

public TrainValidationSplit setEstimatorParamMaps(ParamMap[] value)

setEvaluator

public TrainValidationSplit setEvaluator(Evaluator value)

setTrainRatio

public TrainValidationSplit setTrainRatio(double value)

setSeed

public TrainValidationSplit setSeed(long value)

setParallelism

public TrainValidationSplit setParallelism(int value)

Set the maximum level of parallelism to evaluate models in parallel. Default is 1 for serial evaluation

Parameters:

value - (undocumented)

Returns:

(undocumented)

setCollectSubModels

public TrainValidationSplit setCollectSubModels(boolean value)

Whether to collect submodels when fitting. If set, we can get submodels from the returned model.

Note: If set this param, when you save the returned model, you can set an option "persistSubModels" to be "true" before saving, in order to save these submodels. You can check documents of TrainValidationSplitModel.TrainValidationSplitModelWriter for more information.

Parameters:

value - (undocumented)

Returns:

(undocumented)

fit

public TrainValidationSplitModel fit(Dataset<?> dataset)

Description copied from class: Estimator

Fits a model to the input data.

Specified by:

fit in class Estimator<TrainValidationSplitModel>

Parameters:

dataset - (undocumented)

Returns:

(undocumented)

transformSchema

public StructType transformSchema(StructType schema)

Description copied from class: PipelineStage

:: DeveloperApi ::

Check transform validity and derive the output schema from the input schema.

We check validity for interactions between parameters during transformSchema and raise an exception if any parameter value is invalid. Parameter value checks which do not depend on other parameters are handled by Param.validate().

Typical implementation should first conduct verification on schema change and parameter validity, including complex parameter interaction checks.

Specified by:

transformSchema in class PipelineStage

Parameters:

schema - (undocumented)

Returns:

(undocumented)

copy

public TrainValidationSplit copy(ParamMap extra)

Description copied from interface: Params

Creates a copy of this instance with the same UID and some extra params. Subclasses should implement this method and set the return type properly. See defaultCopy().

Specified by:

copy in interface Params

Specified by:

copy in class Estimator<TrainValidationSplitModel>

Parameters:

extra - (undocumented)

Returns:

(undocumented)

write

public MLWriter write()

Description copied from interface: MLWritable

Returns an MLWriter instance for this ML instance.

Specified by:

write in interface MLWritable

Returns:

(undocumented)

getTrainRatio

public double getTrainRatio()

trainRatio

public DoubleParam trainRatio()

Param for ratio between train and validation data. Must be between 0 and 1. Default: 0.75

Returns:

(undocumented)

estimator

public Param<Estimator<?>> estimator()

param for the estimator to be validated

Returns:

(undocumented)

getEstimator

public Estimator<?> getEstimator()

estimatorParamMaps

public Param<ParamMap[]> estimatorParamMaps()

param for estimator param maps

Returns:

(undocumented)

getEstimatorParamMaps

public ParamMap[] getEstimatorParamMaps()

evaluator

public Param<Evaluator> evaluator()

param for the evaluator used to select hyper-parameters that maximize the validated metric

Returns:

(undocumented)

getEvaluator

public Evaluator getEvaluator()

transformSchemaImpl

public StructType transformSchemaImpl(StructType schema)

logTuningParams

public void logTuningParams(org.apache.spark.ml.util.Instrumentation instrumentation)

Instrumentation logging for tuning params including the inner estimator and evaluator info.

Parameters:

instrumentation - (undocumented)