Packages

c

org.apache.spark.mllib.regression

StreamingLinearRegressionWithSGD

class StreamingLinearRegressionWithSGD extends StreamingLinearAlgorithm[LinearRegressionModel, LinearRegressionWithSGD] with Serializable

Train or predict a linear regression model on streaming data. Training uses Stochastic Gradient Descent to update the model based on each new batch of incoming data from a DStream (see LinearRegressionWithSGD for model equation)

Each batch of data is assumed to be an RDD of LabeledPoints. The number of data points per batch can vary, but the number of features must be constant. An initial weight vector must be provided.

Use a builder pattern to construct a streaming linear regression analysis in an application, like:

val model = new StreamingLinearRegressionWithSGD() .setStepSize(0.5) .setNumIterations(10) .setInitialWeights(Vectors.dense(...)) .trainOn(DStream)

Annotations
@Since("1.1.0")
Source
StreamingLinearRegressionWithSGD.scala
Linear Supertypes
Serializable, StreamingLinearAlgorithm[LinearRegressionModel, LinearRegressionWithSGD], Logging, AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. StreamingLinearRegressionWithSGD
  2. Serializable
  3. StreamingLinearAlgorithm
  4. Logging
  5. AnyRef
  6. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. Protected

Instance Constructors

  1. new StreamingLinearRegressionWithSGD()

    Construct a StreamingLinearRegression object with default parameters: {stepSize: 0.1, numIterations: 50, miniBatchFraction: 1.0}.

    Construct a StreamingLinearRegression object with default parameters: {stepSize: 0.1, numIterations: 50, miniBatchFraction: 1.0}. Initial weights must be set before using trainOn or predictOn (see StreamingLinearAlgorithm)

    Annotations
    @Since("1.1.0")

Type Members

  1. implicit class LogStringContext extends AnyRef
    Definition Classes
    Logging

Value Members

  1. val algorithm: LinearRegressionWithSGD

    The algorithm to use for updating.

    The algorithm to use for updating.

    Definition Classes
    StreamingLinearRegressionWithSGD → StreamingLinearAlgorithm
    Annotations
    @Since("1.1.0")
  2. def latestModel(): LinearRegressionModel

    Return the latest model.

    Return the latest model.

    Definition Classes
    StreamingLinearAlgorithm
    Annotations
    @Since("1.1.0")
  3. def predictOn(data: JavaDStream[Vector]): JavaDStream[Double]

    Java-friendly version of predictOn.

    Java-friendly version of predictOn.

    Definition Classes
    StreamingLinearAlgorithm
    Annotations
    @Since("1.3.0")
  4. def predictOn(data: DStream[Vector]): DStream[Double]

    Use the model to make predictions on batches of data from a DStream

    Use the model to make predictions on batches of data from a DStream

    data

    DStream containing feature vectors

    returns

    DStream containing predictions

    Definition Classes
    StreamingLinearAlgorithm
    Annotations
    @Since("1.1.0")
  5. def predictOnValues[K](data: JavaPairDStream[K, Vector]): JavaPairDStream[K, Double]

    Java-friendly version of predictOnValues.

    Java-friendly version of predictOnValues.

    Definition Classes
    StreamingLinearAlgorithm
    Annotations
    @Since("1.3.0")
  6. def predictOnValues[K](data: DStream[(K, Vector)])(implicit arg0: ClassTag[K]): DStream[(K, Double)]

    Use the model to make predictions on the values of a DStream and carry over its keys.

    Use the model to make predictions on the values of a DStream and carry over its keys.

    K

    key type

    data

    DStream containing feature vectors

    returns

    DStream containing the input keys and the predictions as values

    Definition Classes
    StreamingLinearAlgorithm
    Annotations
    @Since("1.1.0")
  7. def setConvergenceTol(tolerance: Double): StreamingLinearRegressionWithSGD.this.type

    Set the convergence tolerance.

    Set the convergence tolerance. Default: 0.001.

    Annotations
    @Since("1.5.0")
  8. def setInitialWeights(initialWeights: Vector): StreamingLinearRegressionWithSGD.this.type

    Set the initial weights.

    Set the initial weights.

    Annotations
    @Since("1.1.0")
  9. def setMiniBatchFraction(miniBatchFraction: Double): StreamingLinearRegressionWithSGD.this.type

    Set the fraction of each batch to use for updates.

    Set the fraction of each batch to use for updates. Default: 1.0.

    Annotations
    @Since("1.1.0")
  10. def setNumIterations(numIterations: Int): StreamingLinearRegressionWithSGD.this.type

    Set the number of iterations of gradient descent to run per update.

    Set the number of iterations of gradient descent to run per update. Default: 50.

    Annotations
    @Since("1.1.0")
  11. def setRegParam(regParam: Double): StreamingLinearRegressionWithSGD.this.type

    Set the regularization parameter.

    Set the regularization parameter. Default: 0.0.

    Annotations
    @Since("2.0.0")
  12. def setStepSize(stepSize: Double): StreamingLinearRegressionWithSGD.this.type

    Set the step size for gradient descent.

    Set the step size for gradient descent. Default: 0.1.

    Annotations
    @Since("1.1.0")
  13. def trainOn(data: JavaDStream[LabeledPoint]): Unit

    Java-friendly version of trainOn.

    Java-friendly version of trainOn.

    Definition Classes
    StreamingLinearAlgorithm
    Annotations
    @Since("1.3.0")
  14. def trainOn(data: DStream[LabeledPoint]): Unit

    Update the model by training on batches of data from a DStream.

    Update the model by training on batches of data from a DStream. This operation registers a DStream for training the model, and updates the model based on every subsequent batch of data from the stream.

    data

    DStream containing labeled data

    Definition Classes
    StreamingLinearAlgorithm
    Annotations
    @Since("1.1.0")