Evaluator

Type Members

class EmptyEvaluator extends Evaluator[EmptyEvaluator]

Used in case when folding is needed, but not the evaluation
class EvaluatingTransformer[M <: ModelWithSummary[M], E <: Evaluator[E]] extends ModelOnlyTransformer[M, EvaluatingTransformer[M, E]] with HasMetricsBlock

Utility used for transparent injection of the evaluator into training chain.
Utility used for transparent injection of the evaluator into training chain. Evaluator is applied only while fitting (it adds an extra summary block), but has no other traces in the final model (does not affect predictions).
class PostProcessingEvaluator[E <: Evaluator[E]] extends Evaluator[PostProcessingEvaluator[E]]
class TrainOnlyFilter extends Transformer with HasIsTestCol

Utility used to filter out test data before passing to estimator.
class TrainTestEvaluator[N <: Evaluator[N]] extends Evaluator[TrainTestEvaluator[N]] with HasIsTestCol

This is a simple workaround to add kind of grouping by test/train column for evaluators without embedded support for grouping (eg.
This is a simple workaround to add kind of grouping by test/train column for evaluators without embedded support for grouping (eg. BinaryClassificationEvaluator).

Value Members

final def !=(arg0: Any): Boolean

Definition Classes
AnyRef → Any
final def ##(): Int

Definition Classes
AnyRef → Any
final def ==(arg0: Any): Boolean

Definition Classes
AnyRef → Any
def addFolds[M <: ModelWithSummary[M]](estimator: SummarizableEstimator[M], folder: FoldsAssigner = new FoldsAssigner()): UnwrappedStage[M, IdentityModelTransformer[M]]

Adds folds (foldNum column) to the dataset before passing it to the nested estimator.
Adds folds (foldNum column) to the dataset before passing it to the nested estimator.
estimator
Nested predictor for fitting the model
folder
Transformer adding folds (by default based on row hash)
returns
Estimator returning a model fit by nested predictor on a dataset with extra foldNum column
final def asInstanceOf[T0]: T0

Definition Classes
Any
def clone(): AnyRef

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( ... )
def crossValidate[M <: ModelWithSummary[M], E <: Evaluator[E]](estimator: SummarizableEstimator[M], evaluator: E, numFolds: Int = 10, numThreads: Int = 1, prefix: String = "", addGlobal: Boolean = true): SummarizableEstimator[M]

Performs a cross validation given predictor and evaluator.
Performs a cross validation given predictor and evaluator. Returns a model with summary blocks extended with foldNum column.
Split into folds is done based on the hash of entire row.
estimator
Nested predictor for fitting the model.
evaluator
Evaluator for creating a metric.
numFolds
Number of folds for validation (defeult 10)
numThreads
Number of parallel folds training.
returns
Estimator which returns a model fit by the nested predictor on the entire dataset with summary blocks extended with numFolds column.
final def eq(arg0: AnyRef): Boolean

Definition Classes
AnyRef
def equals(arg0: Any): Boolean

Definition Classes
AnyRef → Any
def evaluate[M <: ModelWithSummary[M], E <: Evaluator[E]](estimator: SummarizableEstimator[M], evaluator: E): SummarizableEstimator[M]

Fit and then evaluate model.
Fit and then evaluate model. Results of evaluation is stored into a dedicated summary block.
estimator
Used to fit the model
evaluator
Used to evaluate the model.
returns
Estimator which returns a model fit by nested predictor with extra summary block for metrics, produced by evaluator.
def finalize(): Unit

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( classOf[java.lang.Throwable] )
final def getClass(): Class[_]

Definition Classes
AnyRef → Any
def hashCode(): Int

Definition Classes
AnyRef → Any
final def isInstanceOf[T0]: Boolean

Definition Classes
Any
final def ne(arg0: AnyRef): Boolean

Definition Classes
AnyRef
final def notify(): Unit

Definition Classes
AnyRef
final def notifyAll(): Unit

Definition Classes
AnyRef
final def synchronized[T0](arg0: ⇒ T0): T0

Definition Classes
AnyRef
def toString(): String

Definition Classes
AnyRef → Any
def validateInFolds[M <: ModelWithSummary[M], E <: Evaluator[E]](estimator: SummarizableEstimator[M], evaluator: E, numFolds: Int = 10, numThreads: Int = 1, prefix: String = "", addGlobal: Boolean = true): CrossValidator[M]

Performs a cross validation given predictor and evaluator.
Performs a cross validation given predictor and evaluator. Returns a model with summary blocks extended with foldNum column.
Split into folds is expected to be done externaly.
estimator
Nested predictor for fitting the model.
evaluator
Evaluator for creating a metric.
numFolds
Number of folds for validation (defeult 10)
numThreads
Number of threads to run validation.
returns
Estimator which returns a model fit by the nested predictor on the entire dataset with summary blocks extended with numFolds column.
final def wait(): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )
final def wait(arg0: Long, arg1: Int): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )
final def wait(arg0: Long): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )

Related Docs: class Evaluator | package odkl

object Evaluator extends Serializable

Type Members

class EmptyEvaluator extends Evaluator[EmptyEvaluator]

class EvaluatingTransformer[M <: ModelWithSummary[M], E <: Evaluator[E]] extends ModelOnlyTransformer[M, EvaluatingTransformer[M, E]] with HasMetricsBlock

class PostProcessingEvaluator[E <: Evaluator[E]] extends Evaluator[PostProcessingEvaluator[E]]

class TrainOnlyFilter extends Transformer with HasIsTestCol

class TrainTestEvaluator[N <: Evaluator[N]] extends Evaluator[TrainTestEvaluator[N]] with HasIsTestCol

Value Members

final def !=(arg0: Any): Boolean

final def ##(): Int

final def ==(arg0: Any): Boolean

def addFolds[M <: ModelWithSummary[M]](estimator: SummarizableEstimator[M], folder: FoldsAssigner = new FoldsAssigner()): UnwrappedStage[M, IdentityModelTransformer[M]]

final def asInstanceOf[T0]: T0

def clone(): AnyRef

def crossValidate[M <: ModelWithSummary[M], E <: Evaluator[E]](estimator: SummarizableEstimator[M], evaluator: E, numFolds: Int = 10, numThreads: Int = 1, prefix: String = "", addGlobal: Boolean = true): SummarizableEstimator[M]

final def eq(arg0: AnyRef): Boolean

def equals(arg0: Any): Boolean

def evaluate[M <: ModelWithSummary[M], E <: Evaluator[E]](estimator: SummarizableEstimator[M], evaluator: E): SummarizableEstimator[M]

def finalize(): Unit

final def getClass(): Class[_]

def hashCode(): Int

final def isInstanceOf[T0]: Boolean

final def ne(arg0: AnyRef): Boolean

final def notify(): Unit

final def notifyAll(): Unit

final def synchronized[T0](arg0: ⇒ T0): T0

def toString(): String

def validateInFolds[M <: ModelWithSummary[M], E <: Evaluator[E]](estimator: SummarizableEstimator[M], evaluator: E, numFolds: Int = 10, numThreads: Int = 1, prefix: String = "", addGlobal: Boolean = true): CrossValidator[M]

final def wait(): Unit

final def wait(arg0: Long, arg1: Int): Unit

final def wait(arg0: Long): Unit

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped