All Downloads are FREE. Search and download functionalities are using the official Maven repository.

ai.djl.nn.ParallelBlock Maven / Gradle / Ivy

There is a newer version: 0.30.0
Show newest version
/*
 * Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
 *
 * Licensed under the Apache License, Version 2.0 (the "License"). You may not use this file except in compliance
 * with the License. A copy of the License is located at
 *
 * http://aws.amazon.com/apache2.0/
 *
 * or in the "license" file accompanying this file. This file 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.
 */
package ai.djl.nn;

import ai.djl.MalformedModelException;
import ai.djl.ndarray.NDList;
import ai.djl.ndarray.NDManager;
import ai.djl.ndarray.types.DataType;
import ai.djl.ndarray.types.Shape;
import ai.djl.training.ParameterStore;
import ai.djl.util.PairList;
import java.io.DataInputStream;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collection;
import java.util.Collections;
import java.util.List;
import java.util.function.Function;
import java.util.stream.Collectors;

/**
 * {@code ParallelBlock} is a {@link Block} whose children form a parallel branch in the network and
 * are combined to produce a single output.
 *
 * 

{@code ParallelBlock} has no direct parameters. */ public class ParallelBlock extends AbstractBlock { private static final byte VERSION = 2; private Function, NDList> function; /** * Creates a parallel block whose branches are combined to form a single output by the given * function. * * @param function the function to define how the parallel branches are combined to form a * single output */ public ParallelBlock(Function, NDList> function) { this(function, Collections.emptyList()); } /** * Creates a parallel block whose branches are formed by each block in the list of blocks, and * are combined to form a single output by the given function. * * @param function the function to define how the parallel branches are combined * @param blocks the blocks that form each of the parallel branches */ public ParallelBlock(Function, NDList> function, List blocks) { super(VERSION); this.function = function; addAll(blocks); } /** * Adds an array of blocks, each of which is a parallel branch. * * @param blocks the array of blocks to add * @return this block */ public final ParallelBlock addAll(Block... blocks) { return addAll(Arrays.asList(blocks)); } /** * Adds a {@link Collection} of blocks, each of which is a parallel branch. * * @param blocks the {@link Collection} of blocks to add * @return this block */ public final ParallelBlock addAll(Collection blocks) { blocks.forEach(this::add); return this; } /** * Adds the given {@link Block} to the block, which is one parallel branch. * * @param block the block to be added as a parallel branch * @return this block */ public final ParallelBlock add(Block block) { if (block != null) { addChildBlock(block.getClass().getSimpleName(), block); } return this; } /** * Adds a {@link LambdaBlock}, that applies the given function, to the list of parallel * branches. * * @param f the function that forms the {@link LambdaBlock} * @return this block */ public final ParallelBlock add(Function f) { return add(new LambdaBlock(f)); } /** {@inheritDoc} */ @Override public NDList forward( ParameterStore parameterStore, NDList inputs, boolean training, PairList params) { return function.apply( children.values() .stream() .map(block -> block.forward(parameterStore, inputs, training, params)) .collect(Collectors.toList())); } /** {@inheritDoc} */ @Override public void initializeChildBlocks(NDManager manager, DataType dataType, Shape... inputShapes) { for (Block child : getChildren().values()) { child.initialize(manager, dataType, inputShapes); } } /** {@inheritDoc} */ @Override public Shape[] getOutputShapes(NDManager manager, Shape[] inputShapes) { if (children.isEmpty()) { throw new IllegalArgumentException("The parallel block is empty"); } try (NDManager subManager = manager.newSubManager()) { List inputs = new ArrayList<>(); for (Block block : children.values()) { Shape[] shapes = block.getOutputShapes(manager, inputShapes); NDList output = new NDList(shapes.length); for (Shape shape : shapes) { output.add(subManager.create(shape)); } inputs.add(output); } NDList output = function.apply(inputs); Shape[] outputShapes = new Shape[output.size()]; for (int i = 0; i < output.size(); ++i) { outputShapes[i] = output.get(i).getShape(); } return outputShapes; } } /** {@inheritDoc} */ @Override public void loadMetadata(byte version, DataInputStream is) throws IOException, MalformedModelException { if (version == VERSION) { readInputShapes(is); } else if (version != 1) { throw new MalformedModelException("Unsupported encoding version: " + version); } } /** {@inheritDoc} */ @Override public String toString() { StringBuilder sb = new StringBuilder(200); sb.append("Parallel(\n"); for (Block block : children.values()) { String blockString = block.toString().replaceAll("(?m)^", "\t"); sb.append(blockString).append('\n'); } sb.append(')'); return sb.toString(); } }





© 2015 - 2024 Weber Informatics LLC | Privacy Policy