org.deeplearning4j.spark.impl.multilayer.scoring

Class ScoreExamplesFunction

java.lang.Object

org.deeplearning4j.spark.util.BaseDoubleFlatMapFunctionAdaptee<java.util.Iterator<org.nd4j.linalg.dataset.DataSet>>

org.deeplearning4j.spark.impl.multilayer.scoring.ScoreExamplesFunction

All Implemented Interfaces:

java.io.Serializable, org.apache.spark.api.java.function.DoubleFlatMapFunction<java.util.Iterator<org.nd4j.linalg.dataset.DataSet>>

public class ScoreExamplesFunction
extends BaseDoubleFlatMapFunctionAdaptee<java.util.Iterator<org.nd4j.linalg.dataset.DataSet>>

Function to score examples individually. Note that scoring is batched for computational efficiency.
This is essentially a Spark implementation of the MultiLayerNetwork.scoreExamples(DataSet, boolean) method
Note: This method returns a score for each example, but the association between examples and scores is lost. In cases where we need to know the score for particular examples, use ScoreExamplesWithKeyFunction

See Also:

ScoreExamplesWithKeyFunction, Serialized Form

Field Summary

Fields inherited from class org.deeplearning4j.spark.util.BaseDoubleFlatMapFunctionAdaptee

adapter

Constructor Summary

Constructors

Constructor and Description
ScoreExamplesFunction(org.apache.spark.broadcast.Broadcast<org.nd4j.linalg.api.ndarray.INDArray> params, org.apache.spark.broadcast.Broadcast<java.lang.String> jsonConfig, boolean addRegularizationTerms, int batchSize)

Method Summary

Methods inherited from class org.deeplearning4j.spark.util.BaseDoubleFlatMapFunctionAdaptee

call

Methods inherited from class java.lang.Object

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

Constructor Detail

ScoreExamplesFunction

public ScoreExamplesFunction(org.apache.spark.broadcast.Broadcast<org.nd4j.linalg.api.ndarray.INDArray> params,
                             org.apache.spark.broadcast.Broadcast<java.lang.String> jsonConfig,
                             boolean addRegularizationTerms,
                             int batchSize)