org.nd4j.linalg.api.ops.impl.transforms.clip.ClipByValue Maven / Gradle / Ivy

Show more of this group Show more artifacts with this name
Show all versions of nd4j-api Show documentation
There is a newer version: 1.0.0-M2.1
/*
 *  ******************************************************************************
 *  *
 *  *
 *  * 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.transforms.clip;

import lombok.NonNull;
import onnx.Onnx;
import org.nd4j.autodiff.samediff.SDVariable;
import org.nd4j.autodiff.samediff.SameDiff;
import org.nd4j.common.base.Preconditions;
import org.nd4j.linalg.api.buffer.DataType;
import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.api.ops.DynamicCustomOp;
import org.tensorflow.framework.AttrValue;
import org.tensorflow.framework.GraphDef;
import org.tensorflow.framework.NodeDef;

import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.Map;

public class ClipByValue extends DynamicCustomOp {

    private double clipValueMin;
    private double clipValueMax;

    public ClipByValue(@NonNull INDArray input, double clipValueMin, double clipValueMax) {
        super(null, new INDArray[]{input}, null);
        this.clipValueMin = clipValueMin;
        this.clipValueMax = clipValueMax;
        addTArgument(clipValueMin, clipValueMax);
    }

    public ClipByValue() {

    }

    public ClipByValue(SameDiff sameDiff, SDVariable x, double clipValueMin, double clipValueMax, boolean inPlace) {
        super(null, sameDiff, new SDVariable[]{x});
        this.clipValueMin = clipValueMin;
        this.clipValueMax = clipValueMax;
        this.inplaceCall = inPlace;
        addTArgument(clipValueMin, clipValueMax);
    }


    public ClipByValue(SameDiff sameDiff, SDVariable x, double clipValueMin, double clipValueMax) {
        super(null, sameDiff, new SDVariable[]{x});
        this.clipValueMin = clipValueMin;
        this.clipValueMax = clipValueMax;
        addTArgument(clipValueMin, clipValueMax);
    }

    @Override
    public String opName() {
        return "ClipByValue";
    }

    @Override
    public void initFromTensorFlow(NodeDef nodeDef, SameDiff initWith, Map attributesForNode, GraphDef graph) {
        throw new UnsupportedOperationException("Not yet implemented");
    }

    @Override
    public void initFromOnnx(Onnx.NodeProto node, SameDiff initWith, Map attributesForNode, Onnx.GraphProto graph) {
        throw new UnsupportedOperationException("Not yet implemented");
    }


    @Override
    public List doDiff(List grad) {
        //dOut/dIn is 0 if clipped, 1 otherwise
        SDVariable notClippedLower = sameDiff.gt(arg(), clipValueMin).castTo(arg().dataType());
        SDVariable notClippedUpper = sameDiff.lt(arg(), clipValueMax).castTo(arg().dataType());
        SDVariable ret = notClippedLower.mul(notClippedUpper).mul(grad.get(0));
        return Collections.singletonList(ret);
    }

    @Override
    public List calculateOutputDataTypes(List inputDataTypes){
        Preconditions.checkState(inputDataTypes != null && !inputDataTypes.isEmpty() , "Expected at least 1 input datatype for %s, got %s", getClass(), inputDataTypes);
        //get the final data type (sometimes model import passes in 2 dummy data types that aren't relevant)
        return Arrays.asList(inputDataTypes.get(inputDataTypes.size() - 1));
    }
}