org.deeplearning4j.spark.impl.common.score

Class BaseVaeScoreWithKeyFunctionAdapter<K>

java.lang.Object

org.deeplearning4j.spark.impl.common.score.BaseVaeScoreWithKeyFunctionAdapter<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:

java.io.Serializable, org.datavec.spark.functions.FlatMapFunctionAdapter<java.util.Iterator<scala.Tuple2<K,org.nd4j.linalg.api.ndarray.INDArray>>,scala.Tuple2<K,java.lang.Double>>

Direct Known Subclasses:

BaseVaeReconstructionProbWithKeyFunctionAdapter, CGVaeReconstructionErrorWithKeyFunction

public abstract class BaseVaeScoreWithKeyFunctionAdapter<K>
extends java.lang.Object
implements org.datavec.spark.functions.FlatMapFunctionAdapter<java.util.Iterator<scala.Tuple2<K,org.nd4j.linalg.api.ndarray.INDArray>>,scala.Tuple2<K,java.lang.Double>>

Function to calculate the scores (reconstruction probability, reconstruction error) for a variational autoencoder.
Note that scoring is batched for computational efficiency.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected org.apache.spark.broadcast.Broadcast<java.lang.String>	jsonConfig
protected org.apache.spark.broadcast.Broadcast<org.nd4j.linalg.api.ndarray.INDArray>	params

Constructor Summary

Constructors

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

Method Summary

Modifier and Type	Method and Description
java.lang.Iterable<scala.Tuple2<K,java.lang.Double>>	call(java.util.Iterator<scala.Tuple2<K,org.nd4j.linalg.api.ndarray.INDArray>> iterator)
abstract org.nd4j.linalg.api.ndarray.INDArray	computeScore(VariationalAutoencoder vae, org.nd4j.linalg.api.ndarray.INDArray toScore)
abstract VariationalAutoencoder	getVaeLayer()

Methods inherited from class java.lang.Object

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

Field Detail

params

protected final org.apache.spark.broadcast.Broadcast<org.nd4j.linalg.api.ndarray.INDArray> params

jsonConfig

protected final org.apache.spark.broadcast.Broadcast<java.lang.String> jsonConfig

Constructor Detail

BaseVaeScoreWithKeyFunctionAdapter

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

Parameters:

params - MultiLayerNetwork parameters

jsonConfig - MultiLayerConfiguration, as json

batchSize - Batch size to use when scoring

Method Detail

getVaeLayer

public abstract VariationalAutoencoder getVaeLayer()

computeScore

public abstract org.nd4j.linalg.api.ndarray.INDArray computeScore(VariationalAutoencoder vae,
                                                                  org.nd4j.linalg.api.ndarray.INDArray toScore)

call

public java.lang.Iterable<scala.Tuple2<K,java.lang.Double>> call(java.util.Iterator<scala.Tuple2<K,org.nd4j.linalg.api.ndarray.INDArray>> iterator)
                                                          throws java.lang.Exception

Specified by:

call in interface org.datavec.spark.functions.FlatMapFunctionAdapter<java.util.Iterator<scala.Tuple2<K,org.nd4j.linalg.api.ndarray.INDArray>>,scala.Tuple2<K,java.lang.Double>>

Throws:

java.lang.Exception