org.deeplearning4j.nn.conf.dropout.GaussianNoise Maven / Gradle / Ivy

Go to download
/*
 *  ******************************************************************************
 *  *
 *  *
 *  * 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.nn.conf.dropout;

import lombok.Data;
import org.deeplearning4j.nn.workspace.LayerWorkspaceMgr;
import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.api.ops.impl.transforms.pairwise.arithmetic.AddOp;
import org.nd4j.linalg.api.ops.random.impl.GaussianDistribution;
import org.nd4j.linalg.factory.Nd4j;
import org.nd4j.linalg.schedule.ISchedule;
import org.nd4j.shade.jackson.annotation.JsonProperty;

@Data
public class GaussianNoise implements IDropout {

    private double stddev;
    private ISchedule stddevSchedule;

    /**
     * @param stddev Standard deviation for the mean 0 Gaussian noise
     */
    public GaussianNoise(double stddev){
        this(stddev, null);
    }

    /**
     * @param stddevSchedule Schedule for standard deviation for the mean 0 Gaussian noise
     */
    public GaussianNoise(ISchedule stddevSchedule){
        this(Double.NaN, stddevSchedule);
    }

    protected GaussianNoise(@JsonProperty("stddev") double stddev, @JsonProperty("stddevSchedule") ISchedule stddevSchedule){
        this.stddev = stddev;
        this.stddevSchedule = stddevSchedule;
    }

    @Override
    public INDArray applyDropout(INDArray inputActivations, INDArray output, int iteration, int epoch, LayerWorkspaceMgr workspaceMgr) {
        double currS;
        if(stddevSchedule != null){
            currS = stddevSchedule.valueAt(iteration, epoch);
        } else {
            currS = stddev;
        }

        INDArray noise = Nd4j.createUninitialized(output.dataType(), inputActivations.shape(), inputActivations.ordering());
        Nd4j.getExecutioner().exec(new GaussianDistribution(noise, 0, currS));

        Nd4j.getExecutioner().exec(new AddOp(inputActivations, noise, output));
        return output;
    }

    @Override
    public INDArray backprop(INDArray gradAtOutput, INDArray gradAtInput, int iteration, int epoch) {
        //dL/dIn = dL/dOut * dOut/dIn, with dOut/dIn = 1
        if(gradAtInput == gradAtOutput){
            //Same array (in-place result)
            return gradAtInput;
        } else {
            return gradAtInput.assign(gradAtOutput);
        }
    }

    @Override
    public void clear() {
        //No op
    }

    @Override
    public IDropout clone() {
        return new GaussianNoise(stddev, stddevSchedule);
    }
}