Please wait. This can take some minutes ...
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.
org.nd4j.linalg.api.ops.impl.layers.convolution.Conv2D Maven / Gradle / Ivy
/*
* ******************************************************************************
* *
* *
* * 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.nd4j.linalg.api.ops.impl.layers.convolution;
import lombok.Builder;
import lombok.Getter;
import lombok.NoArgsConstructor;
import lombok.NonNull;
import lombok.extern.slf4j.Slf4j;
import lombok.val;
import onnx.Onnx;
import org.nd4j.autodiff.samediff.SDVariable;
import org.nd4j.autodiff.samediff.SameDiff;
import org.nd4j.common.base.Preconditions;
import org.nd4j.imports.NoOpNameFoundException;
import org.nd4j.imports.converters.DifferentialFunctionClassHolder;
import org.nd4j.imports.descriptors.properties.AttributeAdapter;
import org.nd4j.imports.descriptors.properties.PropertyMapping;
import org.nd4j.imports.descriptors.properties.adapters.*;
import org.nd4j.imports.graphmapper.tf.TFGraphMapper;
import org.nd4j.linalg.api.buffer.DataType;
import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.api.ops.DynamicCustomOp;
import org.nd4j.linalg.api.ops.impl.layers.convolution.config.Conv2DConfig;
import org.nd4j.common.util.ArrayUtil;
import org.tensorflow.framework.AttrValue;
import org.tensorflow.framework.GraphDef;
import org.tensorflow.framework.NodeDef;
import java.lang.reflect.Field;
import java.util.*;
@Slf4j
@Getter
@NoArgsConstructor
public class Conv2D extends DynamicCustomOp {
protected Conv2DConfig config;
private static final String INVALID_CONFIGURATION = "Invalid Conv2D configuration : sW = %s pH = %s dW = %s ";
public Conv2D(@NonNull SameDiff sameDiff, @NonNull SDVariable input, @NonNull SDVariable weights,
SDVariable bias, @NonNull Conv2DConfig conv2DConfig) {
this(sameDiff, wrapFilterNull(input, weights, bias), conv2DConfig);
}
@Builder(builderMethodName = "sameDiffBuilder")
public Conv2D(SameDiff sameDiff,
SDVariable[] inputFunctions,
Conv2DConfig config) {
super(sameDiff, inputFunctions);
initConfig(config);
}
public Conv2D(INDArray[] inputs, INDArray[] outputs, Conv2DConfig config){
super(inputs, outputs);
initConfig(config);
}
public Conv2D(@NonNull INDArray input, @NonNull INDArray weights, INDArray bias, INDArray output, @NonNull Conv2DConfig config){
this(wrapFilterNull(input, weights, bias), wrapOrNull(output), config);
}
public Conv2D(INDArray layerInput, INDArray weights, INDArray bias, Conv2DConfig config) {
this(layerInput, weights, bias, null, config);
}
protected void initConfig(Conv2DConfig config){
this.config = config;
Preconditions.checkState(config.getSW() >= 1 && config.getPH() >= 0 && config.getDW() >= 1,
INVALID_CONFIGURATION,
config.getSH(), config.getPH(), config.getDW());
addArgs();
}
protected void addArgs() {
addIArgument(config.getKH(),
config.getKW(),
config.getSH(),
config.getSW(),
config.getPH(),
config.getPW(),
config.getDH(),
config.getDW(),
ArrayUtil.fromBoolean(config.isSameMode()),
config.getDataFormat().equalsIgnoreCase("NCHW") ? 0 : 1,
config.getWeightsFormat().ordinal());
}
@Override
public long[] iArgs() {
if (iArguments.size() == 0)
addArgs();
return super.iArgs();
}
@Override
public Object getValue(Field property) {
if (config == null) {
config = Conv2DConfig.builder().build();
}
return config.getValue(property);
}
@Override
public Map propertiesForFunction() {
if(config != null)
return config.toProperties();
return Collections.emptyMap();
}
@Override
public void initFromTensorFlow(NodeDef nodeDef, SameDiff initWith, Map attributesForNode, GraphDef graph) {
TFGraphMapper.initFunctionFromProperties(nodeDef.getOp(), this, attributesForNode, nodeDef, graph);
addArgs();
}
@Override
public boolean isConfigProperties() {
return true;
}
@Override
public String configFieldName() {
return "config";
}
@Override
public void initFromOnnx(Onnx.NodeProto node, SameDiff initWith, Map attributesForNode, Onnx.GraphProto graph) {
}
@Override
public Map> attributeAdaptersForFunction() {
Map> ret = new HashMap<>();
Map tfMappings = new LinkedHashMap<>();
val fields = DifferentialFunctionClassHolder.getInstance().getFieldsForFunction(this);
//TF uses [kH, kW, inC, outC] always for weights
tfMappings.put("kH", new NDArrayShapeAdapter(0));
tfMappings.put("kW", new NDArrayShapeAdapter(1));
tfMappings.put("sH", new ConditionalFieldValueIntIndexArrayAdapter("NCHW", 2, 1, fields.get("dataFormat")));
tfMappings.put("sW", new ConditionalFieldValueIntIndexArrayAdapter("NCHW", 3, 2, fields.get("dataFormat")));
tfMappings.put("dH", new ConditionalFieldValueIntIndexArrayAdapter("NCHW", 2, 1, fields.get("dataFormat")));
tfMappings.put("dW", new ConditionalFieldValueIntIndexArrayAdapter("NCHW", 3, 2, fields.get("dataFormat")));
tfMappings.put("isSameMode", new StringEqualsAdapter("SAME"));
Map onnxMappings = new HashMap<>();
onnxMappings.put("kH", new SizeThresholdIntArrayIntIndexAdapter(0, 2, 0));
onnxMappings.put("kW", new SizeThresholdIntArrayIntIndexAdapter(1, 2, 0));
onnxMappings.put("dH", new SizeThresholdIntArrayIntIndexAdapter(0, 2, 0));
onnxMappings.put("dW", new SizeThresholdIntArrayIntIndexAdapter(1, 2, 0));
onnxMappings.put("sH", new SizeThresholdIntArrayIntIndexAdapter(0, 2, 0));
onnxMappings.put("sW", new SizeThresholdIntArrayIntIndexAdapter(1, 2, 0));
onnxMappings.put("isSameMode", new StringEqualsAdapter("SAME"));
ret.put(tensorflowName(), tfMappings);
ret.put(onnxName(), onnxMappings);
return ret;
}
@Override
public Map> mappingsForFunction() {
Map> ret = new HashMap<>();
Map map = new HashMap<>();
val strideMapping = PropertyMapping.builder()
.tfAttrName("strides")
.onnxAttrName("strides")
.propertyNames(new String[]{"sW", "sH"})
.build();
val kernelMappingH = PropertyMapping.builder()
.propertyNames(new String[]{"kH"})
.tfInputPosition(1)
.shapePosition(0)
.onnxAttrName("kernel_shape")
.build();
val kernelMappingW = PropertyMapping.builder()
.propertyNames(new String[]{"kW"})
.tfInputPosition(1)
.shapePosition(1)
.onnxAttrName("kernel_shape")
.build();
val dilationMapping = PropertyMapping.builder()
.onnxAttrName("dilations")
.propertyNames(new String[]{"dW", "dH"})
.tfAttrName("dilations")
.build();
val dataFormat = PropertyMapping.builder()
.onnxAttrName("data_format")
.tfAttrName("data_format")
.propertyNames(new String[]{"dataFormat"})
.build();
val sameMode = PropertyMapping.builder()
.onnxAttrName("auto_pad")
.propertyNames(new String[]{"isSameMode"})
.tfAttrName("padding")
.build();
val paddingWidthHeight = PropertyMapping.builder()
.onnxAttrName("padding")
.propertyNames(new String[]{"pH", "pW"})
.build();
map.put("sW", strideMapping);
map.put("sH", strideMapping);
map.put("kH", kernelMappingH);
map.put("kW", kernelMappingW);
map.put("dW", dilationMapping);
map.put("dH", dilationMapping);
map.put("isSameMode", sameMode);
map.put("pH", paddingWidthHeight);
map.put("pW", paddingWidthHeight);
map.put("dataFormat", dataFormat);
try {
ret.put(onnxName(), map);
} catch (NoOpNameFoundException e) {
//ignore
}
try {
ret.put(tensorflowName(), map);
} catch (NoOpNameFoundException e) {
//ignore
}
return ret;
}
@Override
public String opName() {
return "conv2d";
}
@Override
public List doDiff(List f1) {
List inputs = new ArrayList<>(Arrays.asList(args()));
inputs.add(f1.get(0));
Conv2DDerivative conv2DDerivative = Conv2DDerivative.derivativeBuilder()
.sameDiff(sameDiff)
.config(config)
.inputFunctions(inputs.toArray(new SDVariable[inputs.size()]))
.build();
List ret = Arrays.asList(conv2DDerivative.outputVariables());
return ret;
}
@Override
public String onnxName() {
return "Conv";
}
@Override
public String tensorflowName() {
return "Conv2D";
}
@Override
public String[] tensorflowNames() {
return new String[]{"Conv2D"};
}
@Override
public List calculateOutputDataTypes(List inputDataTypes){
int n = args().length;
Preconditions.checkState(inputDataTypes != null && inputDataTypes.size() == n, "Expected %s input data types for %s, got %s", n, getClass(), inputDataTypes);
return Collections.singletonList(inputDataTypes.get(0));
}
}