org.deeplearning4j.zoo.model.LeNet Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of deeplearning4j-zoo Show documentation
There is a newer version: 1.0.0-M2.1
/*
 *  ******************************************************************************
 *  *
 *  *
 *  * This program and the accompanying materials are made available under the
 *  * terms of the Apache License, Version 2.0 which is available at
 *  * https://www.apache.org/licenses/LICENSE-2.0.
 *  *
 *  *  See the NOTICE file distributed with this work for additional
 *  *  information regarding copyright ownership.
 *  * Unless required by applicable law or agreed to in writing, software
 *  * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
 *  * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
 *  * License for the specific language governing permissions and limitations
 *  * under the License.
 *  *
 *  * SPDX-License-Identifier: Apache-2.0
 *  *****************************************************************************
 */

package org.deeplearning4j.zoo.model;

import lombok.AllArgsConstructor;
import lombok.Builder;
import lombok.NoArgsConstructor;
import org.deeplearning4j.common.resources.DL4JResources;
import org.deeplearning4j.nn.api.Model;
import org.deeplearning4j.nn.api.OptimizationAlgorithm;
import org.deeplearning4j.nn.conf.*;
import org.deeplearning4j.nn.conf.inputs.InputType;
import org.deeplearning4j.nn.conf.layers.ConvolutionLayer;
import org.deeplearning4j.nn.conf.layers.DenseLayer;
import org.deeplearning4j.nn.conf.layers.OutputLayer;
import org.deeplearning4j.nn.conf.layers.SubsamplingLayer;
import org.deeplearning4j.nn.multilayer.MultiLayerNetwork;
import org.deeplearning4j.nn.weights.WeightInit;
import org.deeplearning4j.zoo.ModelMetaData;
import org.deeplearning4j.zoo.PretrainedType;
import org.deeplearning4j.zoo.ZooModel;
import org.deeplearning4j.zoo.ZooType;
import org.nd4j.linalg.activations.Activation;
import org.nd4j.linalg.learning.config.AdaDelta;
import org.nd4j.linalg.learning.config.IUpdater;
import org.nd4j.linalg.lossfunctions.LossFunctions;

@AllArgsConstructor
@Builder
public class LeNet extends ZooModel {

    @Builder.Default private long seed = 1234;
    @Builder.Default private int[] inputShape = new int[] {1, 28, 28};
    @Builder.Default private int numClasses = 0;
    @Builder.Default private IUpdater updater = new AdaDelta();
    @Builder.Default private CacheMode cacheMode = CacheMode.NONE;
    @Builder.Default private WorkspaceMode workspaceMode = WorkspaceMode.ENABLED;
    @Builder.Default private ConvolutionLayer.AlgoMode cudnnAlgoMode = ConvolutionLayer.AlgoMode.PREFER_FASTEST;

    private LeNet() {}

    @Override
    public String pretrainedUrl(PretrainedType pretrainedType) {
        if (pretrainedType == PretrainedType.MNIST)
            return DL4JResources.getURLString("models/lenet_dl4j_mnist_inference.zip");
        else
            return null;
    }

    @Override
    public long pretrainedChecksum(PretrainedType pretrainedType) {
        if (pretrainedType == PretrainedType.MNIST)
            return 1906861161L;
        else
            return 0L;
    }

    @Override
    public Class modelType() {
        return MultiLayerNetwork.class;
    }

    public MultiLayerConfiguration conf() {
        MultiLayerConfiguration conf = new NeuralNetConfiguration.Builder().seed(seed)
                        .activation(Activation.IDENTITY)
                        .weightInit(WeightInit.XAVIER)
                        .optimizationAlgo(OptimizationAlgorithm.STOCHASTIC_GRADIENT_DESCENT)
                        .updater(updater)
                        .cacheMode(cacheMode)
                        .trainingWorkspaceMode(workspaceMode)
                        .inferenceWorkspaceMode(workspaceMode)
                        .cudnnAlgoMode(cudnnAlgoMode)
                        .convolutionMode(ConvolutionMode.Same)
                        .list()
                        // block 1
                        .layer(new ConvolutionLayer.Builder()
                                .name("cnn1")
                                .kernelSize(5, 5)
                                .stride(1, 1)
                                .nIn(inputShape[0])
                                .nOut(20)
                                .activation(Activation.RELU)
                                .build())
                        .layer(new SubsamplingLayer.Builder(SubsamplingLayer.PoolingType.MAX)
                                .name("maxpool1")
                                .kernelSize(2, 2)
                                .stride(2, 2)
                                .build())
                        // block 2
                        .layer(new ConvolutionLayer.Builder()
                                .name("cnn2")
                                .kernelSize(5, 5)
                                .stride(1, 1)
                                .nOut(50)
                                .activation(Activation.RELU).build())
                        .layer(new SubsamplingLayer.Builder(SubsamplingLayer.PoolingType.MAX)
                                .name("maxpool2")
                                .kernelSize(2, 2)
                                .stride(2, 2)
                                .build())
                        // fully connected
                        .layer(new DenseLayer.Builder()
                                .name("ffn1")
                                .activation(Activation.RELU)
                                .nOut(500)
                                .build())
                        // output
                        .layer(new OutputLayer.Builder(LossFunctions.LossFunction.MCXENT)
                                .name("output")
                                .nOut(numClasses)
                                .activation(Activation.SOFTMAX) // radial basis function required
                                .build())
                        .setInputType(InputType.convolutionalFlat(inputShape[2], inputShape[1], inputShape[0]))
                        .build();

        return conf;
    }

    @Override
    public Model init() {
        MultiLayerNetwork network = new MultiLayerNetwork(conf());
        network.init();
        return network;
    }

    @Override
    public ModelMetaData metaData() {
        return new ModelMetaData(new int[][] {inputShape}, 1, ZooType.CNN);
    }

    @Override
    public void setInputShape(int[][] inputShape) {
        this.inputShape = inputShape[0];
    }
}