opennlp.tools.util.featuregen.WindowFeatureGenerator Maven / Gradle / Ivy
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* 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.
*/
package opennlp.tools.util.featuregen;
import java.util.ArrayList;
import java.util.List;
/**
* Generates previous and next features for a given {@link AdaptiveFeatureGenerator}.
* The window size can be specified.
*
* Features:
* Current token is always included unchanged
* Previous tokens are prefixed with p distance
* Next tokens are prefix with n distance
*/
public class WindowFeatureGenerator implements AdaptiveFeatureGenerator {
public static final String PREV_PREFIX = "p";
public static final String NEXT_PREFIX = "n";
private final AdaptiveFeatureGenerator generator;
private final int prevWindowSize;
private final int nextWindowSize;
/**
* Initializes the current instance with the given parameters.
*
* @param generator Feature generator to apply to the window.
* @param prevWindowSize Size of the window to the left of the current token.
* @param nextWindowSize Size of the window to the right of the current token.
*/
public WindowFeatureGenerator(AdaptiveFeatureGenerator generator, int prevWindowSize, int nextWindowSize) {
this.generator = generator;
this.prevWindowSize = prevWindowSize;
this.nextWindowSize = nextWindowSize;
}
/**
* Initializes the current instance with the given parameters.
*
* @param prevWindowSize
* @param nextWindowSize
* @param generators
*/
public WindowFeatureGenerator(int prevWindowSize, int nextWindowSize,
AdaptiveFeatureGenerator... generators) {
this(new AggregatedFeatureGenerator(generators), prevWindowSize, nextWindowSize);
}
/**
* Initializes the current instance. The previous and next window size is 5.
*
* @param generator feature generator
*/
public WindowFeatureGenerator(AdaptiveFeatureGenerator generator) {
this(generator, 5, 5);
}
/**
* Initializes the current instance with the given parameters.
*
* @param generators array of feature generators
*/
public WindowFeatureGenerator(AdaptiveFeatureGenerator... generators) {
this(new AggregatedFeatureGenerator(generators), 5, 5);
}
public void createFeatures(List features, String[] tokens, int index, String[] preds) {
// current features
generator.createFeatures(features, tokens, index, preds);
// previous features
for (int i = 1; i < prevWindowSize + 1; i++) {
if (index - i >= 0) {
List prevFeatures = new ArrayList<>();
generator.createFeatures(prevFeatures, tokens, index - i, preds);
for (String prevFeature : prevFeatures) {
features.add(PREV_PREFIX + i + prevFeature);
}
}
}
// next features
for (int i = 1; i < nextWindowSize + 1; i++) {
if (i + index < tokens.length) {
List nextFeatures = new ArrayList<>();
generator.createFeatures(nextFeatures, tokens, index + i, preds);
for (String nextFeature : nextFeatures) {
features.add(NEXT_PREFIX + i + nextFeature);
}
}
}
}
public void updateAdaptiveData(String[] tokens, String[] outcomes) {
generator.updateAdaptiveData(tokens, outcomes);
}
public void clearAdaptiveData() {
generator.clearAdaptiveData();
}
@Override
public String toString() {
return super.toString() + ": Prev window size: " + prevWindowSize
+ ", Next window size: " + nextWindowSize;
}
}