Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/*
* ******************************************************************************
* *
* *
* * 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.layers;
import org.deeplearning4j.eval.Evaluation;
import org.deeplearning4j.nn.api.layers.IOutputLayer;
import org.deeplearning4j.nn.conf.NeuralNetConfiguration;
import org.deeplearning4j.nn.gradient.DefaultGradient;
import org.deeplearning4j.nn.gradient.Gradient;
import org.deeplearning4j.optimize.Solver;
import org.nd4j.common.base.Preconditions;
import org.nd4j.linalg.api.buffer.DataType;
import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.dataset.api.DataSet;
import org.nd4j.linalg.dataset.api.iterator.DataSetIterator;
import org.nd4j.linalg.factory.Nd4j;
import org.nd4j.linalg.lossfunctions.ILossFunction;
import org.nd4j.common.primitives.Pair;
import org.nd4j.linalg.util.FeatureUtil;
import org.deeplearning4j.nn.workspace.LayerWorkspaceMgr;
import org.deeplearning4j.nn.workspace.ArrayType;
import java.io.Serializable;
import java.util.ArrayList;
import java.util.List;
public class LossLayer extends BaseLayer
implements Serializable, IOutputLayer {
//current input and label matrices
protected INDArray labels;
private transient Solver solver;
private double fullNetworkRegularizationScore;
public LossLayer(NeuralNetConfiguration conf, DataType dataType) {
super(conf, dataType);
}
/** Compute score after labels and input have been set.
* @param fullNetRegTerm Regularization score term for the entire network
* @param training whether score should be calculated at train or test time (this affects things like application of
* dropout, etc)
* @return score (loss function)
*/
@Override
public double computeScore(double fullNetRegTerm, boolean training, LayerWorkspaceMgr workspaceMgr) {
if (input == null || labels == null)
throw new IllegalStateException("Cannot calculate score without input and labels " + layerId());
this.fullNetworkRegularizationScore = fullNetRegTerm;
INDArray preOut = input;
ILossFunction lossFunction = layerConf().getLossFn();
//double score = lossFunction.computeScore(getLabels2d(), preOut, layerConf().getActivationFunction(), maskArray, false);
double score = lossFunction.computeScore(getLabels2d(), preOut, layerConf().getActivationFn(), maskArray,
false);
score /= getInputMiniBatchSize();
score += fullNetworkRegularizationScore;
this.score = score;
return score;
}
/**Compute the score for each example individually, after labels and input have been set.
*
* @param fullNetRegTerm Regularization score term for the entire network (or, 0.0 to not include regularization)
* @return A column INDArray of shape [numExamples,1], where entry i is the score of the ith example
*/
@Override
public INDArray computeScoreForExamples(double fullNetRegTerm, LayerWorkspaceMgr workspaceMgr) {
if (input == null || labels == null)
throw new IllegalStateException("Cannot calculate score without input and labels " + layerId());
INDArray preOut = input;
ILossFunction lossFunction = layerConf().getLossFn();
INDArray scoreArray =
lossFunction.computeScoreArray(getLabels2d(), preOut, layerConf().getActivationFn(), maskArray);
if (fullNetRegTerm != 0.0) {
scoreArray.addi(fullNetRegTerm);
}
return workspaceMgr.leverageTo(ArrayType.ACTIVATIONS, scoreArray);
}
@Override
public void computeGradientAndScore(LayerWorkspaceMgr workspaceMgr) {
if (input == null || labels == null)
return;
INDArray preOut = input;
Pair pair = getGradientsAndDelta(preOut, workspaceMgr);
this.gradient = pair.getFirst();
score = computeScore(fullNetworkRegularizationScore, true, workspaceMgr);
}
@Override
protected void setScoreWithZ(INDArray z) {
throw new RuntimeException("Not supported " + layerId());
}
@Override
public Pair gradientAndScore() {
return new Pair<>(gradient(), score());
}
@Override
public Pair backpropGradient(INDArray epsilon, LayerWorkspaceMgr workspaceMgr) {
return getGradientsAndDelta(input, workspaceMgr);
}
/** Returns tuple: {Gradient,Delta,Output} given preOut */
private Pair getGradientsAndDelta(INDArray preOut, LayerWorkspaceMgr workspaceMgr) {
// delta calculation
ILossFunction lossFunction = layerConf().getLossFn();
INDArray delta = lossFunction.computeGradient(getLabels2d(), preOut, layerConf().getActivationFn(), maskArray);
// grab the empty gradient
Gradient gradient = new DefaultGradient();
delta = workspaceMgr.leverageTo(ArrayType.ACTIVATION_GRAD, delta);
return new Pair<>(gradient, delta);
}
/**
* Gets the gradient from one training iteration
* @return the gradient (bias and weight matrix)
*/
@Override
public Gradient gradient() {
return gradient;
}
@Override
public double calcRegularizationScore(boolean backpropOnlyParams) {
return 0;
}
@Override
public Type type() {
return Type.FEED_FORWARD;
}
@Override
public void fit(INDArray input, LayerWorkspaceMgr workspaceMgr) {
throw new UnsupportedOperationException("Not supported");
}
@Override
public INDArray activate(boolean training, LayerWorkspaceMgr workspaceMgr) {
INDArray z = input;
INDArray ret = layerConf().getActivationFn().getActivation(z.dup(), training);
if (maskArray != null) {
ret.muliColumnVector(maskArray);
}
INDArray out = workspaceMgr.leverageTo(ArrayType.ACTIVATIONS, ret);
return out;
}
@Override
public INDArray activate(INDArray input, boolean training, LayerWorkspaceMgr workspaceMgr) {
setInput(input, workspaceMgr);
return activate(training, workspaceMgr);
}
@Override
public boolean isPretrainLayer() {
return false;
}
@Override
public INDArray params() {
return null;
}
/**
* Sets the input and labels and returns a score for the prediction
* wrt true labels
*
* @param data the data to score
* @return the score for the given input,label pairs
*/
@Override
public double f1Score(DataSet data) {
return f1Score(data.getFeatures(), data.getLabels());
}
/**
* Returns the f1 score for the given examples.
* Think of this to be like a percentage right.
* The higher the number the more it got right.
* This is on a scale from 0 to 1.
*
* @param examples te the examples to classify (one example in each row)
* @param labels the true labels
* @return the scores for each ndarray
*/
@Override
public double f1Score(INDArray examples, INDArray labels) {
Evaluation eval = new Evaluation();
eval.eval(labels, activate(examples, false, LayerWorkspaceMgr.noWorkspacesImmutable()));
return eval.f1();
}
/**
* Returns the number of possible labels
*
* @return the number of possible labels for this classifier
*/
@Override
public int numLabels() {
return (int) labels.size(1);
}
@Override
public void fit(DataSetIterator iter) {
// no-op
}
/**
* Returns the predictions for each example in the dataset
* @param input the matrix to predict
* @return the prediction for the dataset
*/
@Override
public int[] predict(INDArray input) {
INDArray output = activate(input, false, LayerWorkspaceMgr.noWorkspacesImmutable());
Preconditions.checkState(output.rank() == 2, "predict(INDArray) method can only be used on rank 2 output - got array with rank %s", output.rank());
return output.argMax(1).toIntVector();
}
/**
* Return predicted label names
*
* @param dataSet to predict
* @return the predicted labels for the dataSet
*/
@Override
public List predict(DataSet dataSet) {
int[] intRet = predict(dataSet.getFeatures());
List ret = new ArrayList<>();
for (int i : intRet) {
ret.add(i, dataSet.getLabelName(i));
}
return ret;
}
/**
* Fit the model
*
* @param input the examples to classify (one example in each row)
* @param labels the example labels(a binary outcome matrix)
*/
@Override
public void fit(INDArray input, INDArray labels) {
throw new UnsupportedOperationException("LossLayer has no parameters and cannot be fit");
}
/**
* Fit the model
*
* @param data the data to train on
*/
@Override
public void fit(DataSet data) {
fit(data.getFeatures(), data.getLabels());
}
/**
* Fit the model
*
* @param examples the examples to classify (one example in each row)
* @param labels the labels for each example (the number of labels must match
*/
@Override
public void fit(INDArray examples, int[] labels) {
INDArray outcomeMatrix = FeatureUtil.toOutcomeMatrix(labels, numLabels());
fit(examples, outcomeMatrix);
}
@Override
public void clear() {
super.clear();
if (labels != null) {
labels.data().destroy();
labels = null;
}
solver = null;
}
@Override
public INDArray getLabels() {
return labels;
}
@Override
public boolean needsLabels() {
return true;
}
public void setLabels(INDArray labels) {
this.labels = labels;
}
protected INDArray getLabels2d() {
if (labels.rank() > 2) {
return labels.reshape(labels.size(2), labels.size(1));
}
return labels;
}
}