org.nd4j.linalg.api.ops.impl.shape.ExpandDims Maven / Gradle / Ivy
The newest version!
/*
* ******************************************************************************
* *
* *
* * 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.shape;
import lombok.val;
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.descriptors.properties.PropertyMapping;
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.exception.ND4JIllegalArgumentException;
import org.nd4j.linalg.exception.ND4JIllegalStateException;
import org.tensorflow.framework.AttrValue;
import org.tensorflow.framework.GraphDef;
import org.tensorflow.framework.NodeDef;
import java.util.*;
public class ExpandDims extends DynamicCustomOp {
private int jaxis;
public ExpandDims() {
}
public ExpandDims(SameDiff sameDiff, SDVariable args, int axis) {
this(sameDiff, new SDVariable[]{args}, axis);
}
public ExpandDims(SameDiff sameDiff, SDVariable[] args, int axis) {
super(null, sameDiff, args);
if (axis == Integer.MAX_VALUE) {
throw new ND4JIllegalArgumentException("Cannot perform ExpandDims with axis == Integer.MAX_VALUE");
}
this.jaxis = axis;
addIArgument(this.jaxis);
}
public ExpandDims(SameDiff sameDiff, SDVariable[] args) {
super(null, sameDiff, args);
}
public ExpandDims(INDArray[] inputs, INDArray[] outputs) {
super(null, inputs, outputs);
}
public ExpandDims(SameDiff sameDiff, SDVariable[] args, boolean inPlace) {
super(null, sameDiff, args, inPlace);
}
public ExpandDims(INDArray x, int axis){
super(new INDArray[]{x}, null);
this.jaxis = axis;
addIArgument(axis);
}
@Override
public void initFromTensorFlow(NodeDef nodeDef, SameDiff initWith, Map attributesForNode, GraphDef graph) {
val targetNode = TFGraphMapper.getNodeWithNameFromGraph(graph, nodeDef.getInput(1));
val dimArr = TFGraphMapper.getNDArrayFromTensor(targetNode);
if (dimArr != null) {
int axis = dimArr.data().asInt()[0];
this.jaxis = axis;
addIArgument(this.jaxis);
} else {
this.jaxis = Integer.MAX_VALUE;
addIArgument(this.jaxis);
}
}
@Override
public void setPropertiesForFunction(Map properties) {
if(properties.containsKey("axis")) {
Long value = (Long) properties.get("axis");
if(value != null) {
this.jaxis = value.intValue();
}
}
}
@Override
public void configureFromArguments() {
if(!iArguments.isEmpty()) {
this.jaxis = iArguments.get(0).intValue();
}
}
@Override
public Map propertiesForFunction() {
Map ret = new LinkedHashMap<>();
ret.put("axis", axis);
return ret;
}
@Override
public Map> mappingsForFunction() {
Map> ret = new HashMap<>();
val axisMapping = PropertyMapping.builder()
.tfInputPosition(1)
.propertyNames(new String[]{"axis"})
.build();
Map map = new HashMap<>();
map.put("axis", axisMapping);
ret.put(tensorflowName(), map);
return ret;
}
@Override
public void assertValidForExecution() {
val descriptor = getDescriptor();
if (descriptor.getNumInputs() > 0 && numInputArguments() > 2 || numInputArguments() < 1)
throw new ND4JIllegalStateException("Op failure for " + opName() + " Number of inputs is invalid for execution. Specified " + numInputArguments() + " but should be " + descriptor.getNumInputs());
if (descriptor.getNumOutputs() > 0 && numOutputArguments() != descriptor.getNumOutputs())
throw new ND4JIllegalStateException("Op failure for " + opName() + " Number of outputs is invalid for execution. Specified " + numOutputArguments() + " but should be " + descriptor.getNumInputs());
//< 0 means dynamic size
if (descriptor.getNumIArgs() >= 0 && numIArguments() != descriptor.getNumIArgs())
throw new ND4JIllegalStateException("Op failure for " + opName() + " Number of integer arguments is invalid for execution. Specified " + numIArguments() + " but should be " + descriptor.getNumIArgs());
if (descriptor.getNumTArgs() >= 0 && numTArguments() != descriptor.getNumTArgs())
throw new ND4JIllegalStateException("Op failure for " + opName() + " Number of inputs is invalid for execution. Specified " + numTArguments() + " but should be " + descriptor.getNumTArgs());
}
@Override
public String opName() {
return "expand_dims";
}
@Override
public String onnxName() {
throw new NoOpNameFoundException("No onnx op opName found for " + opName());
}
@Override
public String tensorflowName() {
return "ExpandDims";
}
@Override
public List doDiff(List i_v) {
//Simply need a reshape to remove the dimension...
SDVariable ret = sameDiff.squeeze(i_v.get(0), jaxis);
return Arrays.asList(ret);
}
@Override
public List calculateOutputDataTypes(List dataTypes){
//Axis may be defined either as integer or as an array
Preconditions.checkState(dataTypes != null && (dataTypes.size() == 1 || dataTypes.size() == 2), "Expected list with 1 or 2 datatype for %s, got %s", getClass(), dataTypes);
//Output type is same as input type
return Collections.singletonList(dataTypes.get(0));
}
}