org.deeplearning4j.zoo.model

Class NASNet

java.lang.Object

org.deeplearning4j.zoo.ZooModel

org.deeplearning4j.zoo.model.NASNet

All Implemented Interfaces:

InstantiableModel

public class NASNet
extends ZooModel

U-Net Implementation of NASNet-A in Deeplearning4j. NASNet refers to Neural Architecture Search Network, a family of models that were designed automatically by learning the model architectures directly on the dataset of interest.

This implementation uses 1056 penultimate filters and an input shape of (3, 224, 224). You can change this.

Paper: https://arxiv.org/abs/1707.07012

ImageNet weights for this model are available and have been converted from https://keras.io/applications/.

Author:

Justin Long (crockpotveggies)

Method Summary

Modifier and Type	Method and Description
ComputationGraphConfiguration.GraphBuilder	graphBuilder()
ComputationGraph	init()
ModelMetaData	metaData()
Class<? extends Model>	modelType()
long	pretrainedChecksum(PretrainedType pretrainedType)
String	pretrainedUrl(PretrainedType pretrainedType)
void	setInputShape(int[][] inputShape)

Methods inherited from class org.deeplearning4j.zoo.ZooModel

initPretrained, initPretrained, modelName, pretrainedAvailable

Methods inherited from class java.lang.Object

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

Method Detail

pretrainedUrl

public String pretrainedUrl(PretrainedType pretrainedType)

pretrainedChecksum

public long pretrainedChecksum(PretrainedType pretrainedType)

modelType

public Class<? extends Model> modelType()

init

public ComputationGraph init()

graphBuilder

public ComputationGraphConfiguration.GraphBuilder graphBuilder()

metaData

public ModelMetaData metaData()

setInputShape

public void setInputShape(int[][] inputShape)