deeplearning4j/deeplearning4j-nn/src/main/java/org/deeplearning4j/nn/layers/convolution/Deconvolution2DLayer.java
/*
* ******************************************************************************
* *
* *
* * 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.convolution;
import lombok.val;
import org.deeplearning4j.exception.DL4JInvalidInputException;
import org.deeplearning4j.nn.conf.CNN2DFormat;
import org.deeplearning4j.nn.conf.CacheMode;
import org.deeplearning4j.nn.conf.ConvolutionMode;
import org.deeplearning4j.nn.conf.NeuralNetConfiguration;
import org.deeplearning4j.nn.gradient.DefaultGradient;
import org.deeplearning4j.nn.gradient.Gradient;
import org.deeplearning4j.nn.params.DeconvolutionParamInitializer;
import org.deeplearning4j.util.ConvolutionUtils;
import org.nd4j.linalg.activations.IActivation;
import org.nd4j.linalg.api.buffer.DataType;
import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.api.ops.CustomOp;
import org.nd4j.linalg.api.ops.DynamicCustomOp;
import org.nd4j.linalg.api.shape.Shape;
import org.nd4j.linalg.factory.Nd4j;
import org.nd4j.common.primitives.Pair;
import org.deeplearning4j.nn.workspace.LayerWorkspaceMgr;
import org.deeplearning4j.nn.workspace.ArrayType;
import java.util.Arrays;
public class Deconvolution2DLayer extends ConvolutionLayer {
public Deconvolution2DLayer(NeuralNetConfiguration conf, DataType dataType) {
super(conf, dataType);
}
@Override
public Pair<Gradient, INDArray> backpropGradient(INDArray epsilon, LayerWorkspaceMgr workspaceMgr) {
assertInputSet(true);
if (input.rank() != 4) {
throw new DL4JInvalidInputException("Got rank " + input.rank()
+ " array as input to Deconvolution2DLayer with shape " + Arrays.toString(input.shape())
+ ". Expected rank 4 array with shape " + layerConf().getCnn2dDataFormat().dimensionNames() + ". "
+ layerId());
}
INDArray weights = getParamWithNoise(DeconvolutionParamInitializer.WEIGHT_KEY, true, workspaceMgr);
CNN2DFormat format = layerConf().getCnn2dDataFormat();
boolean nchw = format == CNN2DFormat.NCHW;
int hDim = nchw ? 2 : 1;
int wDim = nchw ? 3 : 2;
long miniBatch = input.size(0);
long inH = input.size(hDim);
long inW = input.size(wDim);
long inDepth = weights.size(0);
long kH = weights.size(2);
long kW = weights.size(3);
int[] dilation = layerConf().getDilation();
int[] kernel = layerConf().getKernelSize();
int[] strides = layerConf().getStride();
int[] pad;
if (convolutionMode == ConvolutionMode.Same) {
int[] outSize = new int[]{(int)epsilon.size(hDim), (int)epsilon.size(wDim)};
pad = ConvolutionUtils.getSameModeTopLeftPadding(outSize, new int[] {(int)inH, (int)inW}, kernel, strides, dilation);
} else {
pad = layerConf().getPadding();
}
INDArray biasGradView = gradientViews.get(DeconvolutionParamInitializer.BIAS_KEY);
INDArray weightGradView = gradientViews.get(DeconvolutionParamInitializer.WEIGHT_KEY);
long[] epsShape = nchw ? new long[]{miniBatch, inDepth, inH, inW} : new long[]{miniBatch, inH, inW, inDepth};
INDArray outEps = workspaceMgr.create(ArrayType.ACTIVATION_GRAD, weights.dataType(), epsShape, 'c');
Integer sameMode = (convolutionMode == ConvolutionMode.Same) ? 1 : 0;
int[] args = new int[] {
(int)kH, (int)kW, strides[0], strides[1],
pad[0], pad[1], dilation[0], dilation[1], sameMode,
nchw ? 0 : 1 //0 = NCHW; 1 = NHWC
};
INDArray delta;
IActivation afn = layerConf().getActivationFn();
Pair<INDArray, INDArray> p = preOutput4d(true, true, workspaceMgr);
delta = afn.backprop(p.getFirst(), epsilon).getFirst();
//DL4J Deconv weights: [inputDepth, outputDepth, kH, kW]
//libnd4j weights: [kH, kW, oC, iC]
weights = weights.permute(2, 3, 1, 0);
INDArray weightGradViewOp = weightGradView.permute(2, 3, 1, 0);
INDArray[] opInputs;
INDArray[] opOutputs;
if(layerConf().hasBias()){
INDArray bias = getParamWithNoise(DeconvolutionParamInitializer.BIAS_KEY, true, workspaceMgr);
opInputs = new INDArray[]{input, weights, bias, delta};
opOutputs = new INDArray[]{outEps, weightGradViewOp, biasGradView};
} else {
opInputs = new INDArray[]{input, weights, delta};
opOutputs = new INDArray[]{outEps, weightGradViewOp};
}
CustomOp op = DynamicCustomOp.builder("deconv2d_bp")
.addInputs(opInputs)
.addIntegerArguments(args)
.addOutputs(opOutputs)
.callInplace(false)
.build();
Nd4j.getExecutioner().exec(op);
Gradient retGradient = new DefaultGradient();
if(layerConf().hasBias()){
retGradient.setGradientFor(DeconvolutionParamInitializer.BIAS_KEY, biasGradView);
}
retGradient.setGradientFor(DeconvolutionParamInitializer.WEIGHT_KEY, weightGradView, 'c');
weightNoiseParams.clear();
return new Pair<>(retGradient, outEps);
}
@Override
protected Pair<INDArray, INDArray> preOutput(boolean training , boolean forBackprop, LayerWorkspaceMgr workspaceMgr) {
INDArray bias = getParamWithNoise(DeconvolutionParamInitializer.BIAS_KEY, training, workspaceMgr);
INDArray weights = getParamWithNoise(DeconvolutionParamInitializer.WEIGHT_KEY, training, workspaceMgr);
//Input validation: expect rank 4 matrix
if (input.rank() != 4) {
String layerName = conf.getLayer().getLayerName();
if (layerName == null)
layerName = "(not named)";
throw new DL4JInvalidInputException("Got rank " + input.rank()
+ " array as input to Deconvolution2D (layer name = " + layerName + ", layer index = "
+ index + ") with shape " + Arrays.toString(input.shape()) + ". "
+ "Expected rank 4 array with shape [minibatchSize, layerInputDepth, inputHeight, inputWidth]."
+ (input.rank() == 2
? " (Wrong input type (see InputType.convolutionalFlat()) or wrong data type?)"
: "")
+ " " + layerId());
}
CNN2DFormat format = layerConf().getCnn2dDataFormat();
boolean nchw = format == CNN2DFormat.NCHW;
int cDim = nchw ? 1 : 3;
int hDim = nchw ? 2 : 1;
int wDim = nchw ? 3 : 2;
long inDepth = weights.size(0);
long outDepth = weights.size(1);
if (input.size(cDim) != inDepth ) {
String layerName = conf.getLayer().getLayerName();
if (layerName == null)
layerName = "(not named)";
String s = "Cannot do forward pass in Deconvolution2D layer (layer name = " + layerName
+ ", layer index = " + index + "): input array channels does not match CNN layer configuration"
+ " (data format = " + format + ", data input channels = " + input.size(cDim) + ", "
+ (nchw ? "[minibatch,inputDepth,height,width]" : "[minibatch,height,width,inputDepth]") + "="
+ Arrays.toString(input.shape()) + "; expected" + " input channels = " + inDepth + ") "
+ layerId();
int dimIfWrongFormat = format == CNN2DFormat.NHWC ? 1 : 3;
if(input.size(dimIfWrongFormat) == inDepth){
//User might have passed NCHW data to a NHWC net, or vice versa?
s += "\n" + ConvolutionUtils.NCHW_NHWC_ERROR_MSG;
}
throw new DL4JInvalidInputException(s);
}
int kH = (int) weights.size(2);
int kW = (int) weights.size(3);
int[] dilation = layerConf().getDilation();
int[] kernel = layerConf().getKernelSize();
int[] strides = layerConf().getStride();
int[] pad;
int[] outSize;
if (convolutionMode == ConvolutionMode.Same) {
outSize = ConvolutionUtils.getDeconvolutionOutputSize(input, kernel, strides, null, convolutionMode, dilation, format); //Also performs validation
pad = ConvolutionUtils.getSameModeTopLeftPadding(outSize, new int[] {(int) input.size(hDim), (int) input.size(wDim)}, kernel,
strides, dilation );
} else {
pad = layerConf().getPadding();
outSize = ConvolutionUtils.getDeconvolutionOutputSize(input, kernel, strides, pad, convolutionMode, dilation, format); //Also performs validation
}
long outH = outSize[0];
long outW = outSize[1];
val miniBatch = input.size(0);
long[] outShape = nchw ? new long[]{miniBatch, outDepth, outH, outW} : new long[]{miniBatch, outH, outW, outDepth};
INDArray output = workspaceMgr.create(ArrayType.ACTIVATIONS, input.dataType(), outShape, 'c');
int sameMode = (convolutionMode == ConvolutionMode.Same) ? 1 : 0;
int[] args = new int[] {
kH, kW, strides[0], strides[1],
pad[0], pad[1], dilation[0], dilation[1], sameMode,
nchw ? 0 : 1 //0 = NCHW; 1 = NHWC
};
//DL4J Deconv weights: [inputDepth, outputDepth, kH, kW]
//libnd4j weights: [kH, kW, oC, iC]
weights = weights.permute(2, 3, 1, 0);
INDArray[] opInputs;
if (layerConf().hasBias()) {
opInputs = new INDArray[]{input, weights, bias};
} else {
opInputs = new INDArray[]{input, weights};
}
CustomOp op = DynamicCustomOp.builder("deconv2d")
.addInputs(opInputs)
.addIntegerArguments(args)
.addOutputs(output)
.callInplace(false)
.build();
Nd4j.getExecutioner().exec(op);
return new Pair<>(output, null);
}
@Override
public INDArray activate(boolean training, LayerWorkspaceMgr workspaceMgr) {
assertInputSet(false);
if (cacheMode == null)
cacheMode = CacheMode.NONE;
applyDropOutIfNecessary(training, workspaceMgr);
INDArray z = preOutput(training, false, workspaceMgr).getFirst();
IActivation afn = layerConf().getActivationFn();
if (helper != null && Shape.strideDescendingCAscendingF(z)) {
INDArray ret = helper.activate(z, layerConf().getActivationFn(), training);
if (ret != null) {
return ret;
}
}
INDArray activation = afn.getActivation(z, training);
return activation;
}
}