weka.dl4j.zoo

Class CustomNet

java.lang.Object

weka.dl4j.zoo.AbstractZooModel

weka.dl4j.zoo.CustomNet

All Implemented Interfaces:

java.io.Serializable, weka.core.OptionHandler

public class CustomNet
extends AbstractZooModel

A dummy ZooModel which is empty.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
CustomNet()

Method Summary

Modifier and Type	Method and Description
org.nd4j.linalg.dataset.api.preprocessor.ImagePreProcessingScaler	getImagePreprocessingScaler() By default, training a model uses the below scaler for the Dl4jMlpClassifier.
int[]	getInputShape() Get the input shape of this zoomodel
java.lang.Enum	getVariation() Get the current variation of the zoo model (e.g., Resnet50 or Resnet101)
org.deeplearning4j.nn.graph.ComputationGraph	init(int numLabels, long seed, int[] shape, boolean filterMode) Initialize the ZooModel as MLP.

Methods inherited from class weka.dl4j.zoo.AbstractZooModel

getChannelsLast, getDefaultGraph, getExtraLayersToRemove, getFeatureExtractionLayer, getNumFExtractOutputs, getOptions, getOutputlayer, getPretrainedType, getPrettyName, initZooModel, isPretrained, isRequiresPooling, listOptions, requiresPreProcessing, setChannelsLast, setExtraLayersToRemove, setFeatureExtractionLayer, setNumFExtractOutputs, setOptions, setOutputLayer, setPretrainedType, setRequiresPooling

Methods inherited from class java.lang.Object

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

Methods inherited from interface weka.core.OptionHandler

makeCopy

Constructor Detail

CustomNet

public CustomNet()

Method Detail

init

public org.deeplearning4j.nn.graph.ComputationGraph init(int numLabels,
                                                         long seed,
                                                         int[] shape,
                                                         boolean filterMode)
                                                  throws java.lang.UnsupportedOperationException

Description copied from class: AbstractZooModel

Initialize the ZooModel as MLP.

Specified by:

init in class AbstractZooModel

Parameters:

numLabels - Number of labels to adjust the output

seed - Seed

shape - shape

filterMode - True if creating for feature extraction

Returns:

MultiLayerNetwork of the specified ZooModel

Throws:

java.lang.UnsupportedOperationException - Init(...) was not supported (only CustomNet)

getInputShape

public int[] getInputShape()

Description copied from class: AbstractZooModel

Get the input shape of this zoomodel

Specified by:

getInputShape in class AbstractZooModel

Returns:

Input shape of this zoomodel

getVariation

public java.lang.Enum getVariation()

Get the current variation of the zoo model (e.g., Resnet50 or Resnet101)

Specified by:

getVariation in class AbstractZooModel

Returns:

Variation

getImagePreprocessingScaler

public org.nd4j.linalg.dataset.api.preprocessor.ImagePreProcessingScaler getImagePreprocessingScaler()

By default, training a model uses the below scaler for the Dl4jMlpClassifier. We also define it here for use in the Inference panel

Specified by:

getImagePreprocessingScaler in class AbstractZooModel

Returns:

Preprocessing scaler used during training.