NNEstimator

final def !=(arg0: Any): Boolean

final def ##(): Int

final def $[T](param: Param[T]): T

final def ==(arg0: Any): Boolean

final def asInstanceOf[T0]: T0

final val batchSize: IntParam

Global batch size across the cluster.

final val cachingSample: BooleanParam

whether to cache the Samples after preprocessing.

final val checkpointOverwrite: BooleanParam

final val checkpointPath: Param[String]

Set a check point saved at path triggered by trigger Default: not enabled

final val checkpointTrigger: Param[Trigger]

final def clear(param: Param[_]): NNEstimator.this.type

def clearGradientClipping(): NNEstimator.this.type

Clear clipping params, in this case, clipping will not be applied.

def clone(): AnyRef

final val constantGradientClippingParams: Param[(Float, Float)]

Constant gradient clipping thresholds.

def copy(extra: ParamMap): NNEstimator[T]

Return a deep copy for DLEstimator.

def copyValues[T <: Params](to: T, extra: ParamMap): T

val criterion: Criterion[T]

BigDL criterion

final def defaultCopy[T <: Params](extra: ParamMap): T

final val endWhen: Param[Trigger]

When to stop the training, passed in a Trigger.

final def eq(arg0: AnyRef): Boolean

def equals(arg0: Any): Boolean

def explainParam(param: Param[_]): String

def explainParams(): String

final def extractParamMap(): ParamMap

final def extractParamMap(extra: ParamMap): ParamMap

final val featuresCol: Param[String]

def finalize(): Unit

def fit(dataset: Dataset[_]): NNModel[T]

def fit(dataset: Dataset[_], paramMaps: Array[ParamMap]): Seq[NNModel[T]]

def fit(dataset: Dataset[_], paramMap: ParamMap): NNModel[T]

def fit(dataset: Dataset[_], firstParamPair: ParamPair[_], otherParamPairs: ParamPair[_]*): NNModel[T]

final def get[T](param: Param[T]): Option[T]

def getBatchSize: Int

def getCheckpointPath: String

Get check point path.

final def getClass(): Class[_]

final def getDefault[T](param: Param[T]): Option[T]

def getEndWhen: Trigger

final def getFeaturesCol: String

final def getLabelCol: String

def getLearningRate: Double

def getLearningRateDecay: Double

def getMaxEpoch: Int

def getOptimMethod: OptimMethod[T]

final def getOrDefault[T](param: Param[T]): T

def getParam(paramName: String): Param[Any]

final def getPredictionCol: String

def getSamplePreprocessing: Preprocessing[Any, Sample[T]]

def getTrainSummary: Option[TrainSummary]

def getValidation: Option[(Trigger, DataFrame, Array[ValidationMethod[T]], Int)]

get the validate configuration during training

def getValidationSummary: Option[ValidationSummary]

Statistics (LearningRate, Loss, Throughput, Parameters) collected during training for the validation data if validation data is set, which can be used for visualization via Tensorboard.

def getVectorSeq(row: Row, colType: DataType, index: Int): Seq[AnyVal]

final def hasDefault[T](param: Param[T]): Boolean

def hasParam(paramName: String): Boolean

def hashCode(): Int

def initializeLogIfNecessary(isInterpreter: Boolean): Unit

def internalFit(dataFrame: DataFrame): NNModel[T]

def isCachingSample: Boolean

final def isDefined(param: Param[_]): Boolean

final def isInstanceOf[T0]: Boolean

final def isSet(param: Param[_]): Boolean

def isTraceEnabled(): Boolean

final val l2GradientClippingParams: FloatParam

L2 norm gradient clipping threshold.

final val labelCol: Param[String]

def log: Logger

def logDebug(msg: ⇒ String, throwable: Throwable): Unit

def logDebug(msg: ⇒ String): Unit

def logError(msg: ⇒ String, throwable: Throwable): Unit

def logError(msg: ⇒ String): Unit

def logInfo(msg: ⇒ String, throwable: Throwable): Unit

def logInfo(msg: ⇒ String): Unit

def logName: String

def logTrace(msg: ⇒ String, throwable: Throwable): Unit

def logTrace(msg: ⇒ String): Unit

def logWarning(msg: ⇒ String, throwable: Throwable): Unit

def logWarning(msg: ⇒ String): Unit

final val maxEpoch: IntParam

Number of max Epoch for the training, an epoch refers to a traverse over the training data Default: 50

val model: Module[T]

BigDL module to be optimized

final def ne(arg0: AnyRef): Boolean

final def notify(): Unit

final def notifyAll(): Unit

final val optimMethod: Param[OptimMethod[T]]

optimization method to be used.

lazy val params: Array[Param[_]]

final val predictionCol: Param[String]

final val samplePreprocessing: Param[Preprocessing[Any, Sample[T]]]

final def set(paramPair: ParamPair[_]): NNEstimator.this.type

final def set(param: String, value: Any): NNEstimator.this.type

final def set[T](param: Param[T], value: T): NNEstimator.this.type

def setBatchSize(value: Int): NNEstimator.this.type

Set global batch size across the cluster.

def setCachingSample(value: Boolean): NNEstimator.this.type

def setCheckpoint(path: String, trigger: Trigger, isOverWrite: Boolean = true): NNEstimator.this.type

Set check points during training.

def setConstantGradientClipping(min: Float, max: Float): NNEstimator.this.type

final def setDefault(paramPairs: ParamPair[_]*): NNEstimator.this.type

final def setDefault[T](param: Param[T], value: T): NNEstimator.this.type

def setEndWhen(trigger: Trigger): NNEstimator.this.type

def setFeaturesCol(featuresColName: String): NNEstimator.this.type

def setGradientClippingByL2Norm(clipNorm: Float): NNEstimator.this.type

def setLabelCol(labelColName: String): NNEstimator.this.type

def setMaxEpoch(value: Int): NNEstimator.this.type

def setOptimMethod(value: OptimMethod[T]): NNEstimator.this.type

def setPredictionCol(value: String): NNEstimator.this.type

def setSamplePreprocessing[FF, LL](value: Preprocessing[(FF, Option[LL]), Sample[T]]): NNEstimator.this.type

def setTrainSummary(value: TrainSummary): NNEstimator.this.type

Statistics (LearningRate, Loss, Throughput, Parameters) collected during training for the training data, which can be used for visualization via Tensorboard.

def setValidation(trigger: Trigger, validationDF: DataFrame, vMethods: Array[ValidationMethod[T]], batchSize: Int): NNEstimator.this.type

Set a validate evaluation during training.

def setValidationSummary(value: ValidationSummary): NNEstimator.this.type

Enable validation Summary.

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

def toString(): String

def transformSchema(schema: StructType): StructType

def transformSchema(schema: StructType, logging: Boolean): StructType

val uid: String

def unwrapVectorAsNecessary(colType: DataType): (Row, Int) ⇒ Any

val validVectorTypes: Seq[UserDefinedType[_ >: Vector with Vector <: Serializable] { def sqlType: org.apache.spark.sql.types.StructType }]

def validateParams(schema: StructType): Unit

var validationBatchSize: Int

var validationDF: DataFrame

var validationMethods: Array[ValidationMethod[T]]

var validationTrigger: Option[Trigger]

final def wait(): Unit

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

final def wait(arg0: Long): Unit

def wrapBigDLModel(m: Module[T]): NNModel[T]

sub classes can extend the method and return required model for different transform tasks

final val learningRate: DoubleParam

learning rate for the optimizer in the NNEstimator.

final val learningRateDecay: DoubleParam

learning rate decay for each iteration.

def setLearningRate(value: Double): NNEstimator.this.type

:: deprecated, please set the learning rate with optimMethod directly.

def setLearningRateDecay(value: Double): NNEstimator.this.type

:: deprecated, please set with optimMethod directly.