org.deeplearning4j.spark.impl.computationgraph.scoring

Class ScoreExamplesWithKeyFunction<K>

java.lang.Object

org.deeplearning4j.spark.impl.computationgraph.scoring.ScoreExamplesWithKeyFunction<K>

Type Parameters:

K - Type of key, associated with each example. Used to keep track of which score belongs to which example

All Implemented Interfaces:

Serializable, org.apache.spark.api.java.function.PairFlatMapFunction<Iterator<scala.Tuple2<K,org.nd4j.linalg.dataset.api.MultiDataSet>>,K,Double>

public class ScoreExamplesWithKeyFunction<K>
extends Object
implements org.apache.spark.api.java.function.PairFlatMapFunction<Iterator<scala.Tuple2<K,org.nd4j.linalg.dataset.api.MultiDataSet>>,K,Double>

Function to score examples individually, where each example is associated with a particular key
Note that scoring is batched for computational efficiency.
This is the Spark implementation of the ComputationGraph.scoreExamples(MultiDataSet, boolean) method
Note: The MultiDataSet objects passed in must have exactly one example in them (otherwise: can't have a 1:1 association between keys and data sets to score)

Author:

Alex Black

See Also:

ScoreExamplesFunction, Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected static org.slf4j.Logger	log

Constructor Summary

Constructors

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

Method Summary

Modifier and Type	Method and Description
Iterable<scala.Tuple2<K,Double>>	call(Iterator<scala.Tuple2<K,org.nd4j.linalg.dataset.api.MultiDataSet>> iterator)

Methods inherited from class java.lang.Object

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

Field Detail

log

protected static org.slf4j.Logger log

Constructor Detail

ScoreExamplesWithKeyFunction

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

Parameters:

params - ComputationGraph parameters

jsonConfig - ComputationGraphConfiguration, as json

addRegularizationTerms - if true: add regularization terms (l1/l2) if applicable; false: don't add regularization terms

batchSize - Batch size to use when scoring examples

Method Detail

call

public Iterable<scala.Tuple2<K,Double>> call(Iterator<scala.Tuple2<K,org.nd4j.linalg.dataset.api.MultiDataSet>> iterator)
                                      throws Exception

Specified by:

call in interface org.apache.spark.api.java.function.PairFlatMapFunction<Iterator<scala.Tuple2<K,org.nd4j.linalg.dataset.api.MultiDataSet>>,K,Double>

Throws:

Exception