/**
* Copyright (c) 2014, the LESK-WSD-DSM AUTHORS.
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* Redistributions of source code must retain the above copyright notice, this
* list of conditions and the following disclaimer.
*
* Redistributions in binary form must reproduce the above copyright notice,
* this list of conditions and the following disclaimer in the documentation
* and/or other materials provided with the distribution.
*
* Neither the name of the University of Bari nor the names of its contributors
* may be used to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*
* GNU GENERAL PUBLIC LICENSE - Version 3, 29 June 2007
*
*/
package di.uniba.it.wsd.dsm;
import java.io.File;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import java.util.PriorityQueue;
import java.util.Set;
import java.util.logging.Level;
import java.util.logging.Logger;
import org.apache.lucene.store.FSDirectory;
import org.apache.lucene.store.IndexInput;
/**
* This class reads and stores in a Map the word vectors
*
* @author pierpaolo
*/
public class LuceneVectorStore implements VectorStore {
private IndexInput indexInput;
private final Map<String, float[]> memory = new HashMap<>();
/**
*
* @param file
* @throws IOException
*/
public void init(File file) throws IOException {
FSDirectory dir = FSDirectory.open(file.getParentFile());
this.indexInput = dir.openInput(file.getName());
String header = indexInput.readString(); //skip header
if ((header.equalsIgnoreCase("-dimensions"))) {
ObjectVector.vecLength = indexInput.readInt();
} else if (header.contains("-dimension")) {
int index = header.indexOf("-dimension");
ObjectVector.vecLength = Integer.parseInt(header.substring(index + 10).trim());
}
loadInRam();
}
/**
*
* @throws IOException
*/
private void loadInRam() throws IOException {
this.indexInput.seek(0);
memory.clear();
String header = indexInput.readString(); //skip header
if ((header.equalsIgnoreCase("-dimensions"))) {
ObjectVector.vecLength = indexInput.readInt();
} else if (header.contains("-dimension")) {
int index = header.indexOf("-dimension");
ObjectVector.vecLength = Integer.parseInt(header.substring(index + 10).trim());
}
while (indexInput.getFilePointer() < indexInput.length()) {
String term = indexInput.readString();
float[] v = new float[ObjectVector.vecLength];
for (int k = 0; k < v.length; k++) {
v[k] = Float.intBitsToFloat(indexInput.readInt());
}
memory.put(term, v);
}
Logger.getLogger(LuceneVectorStore.class.getName()).log(Level.INFO, "Loaded {0} vectors", memory.size());
}
/**
*
* @param set
* @return
* @throws IOException
*/
public Map<String, float[]> prefetch(Set<String> set) throws IOException {
Logger.getLogger(LuceneVectorStore.class.getName()).log(Level.INFO, "Prefetching for {0} vectors", set.size());
this.indexInput.seek(0);
Map<String, float[]> map = new HashMap<>();
String header = indexInput.readString(); //skip header
if ((header.equalsIgnoreCase("-dimensions"))) {
ObjectVector.vecLength = indexInput.readInt();
} else if (header.contains("-dimension")) {
int index = header.indexOf("-dimension");
ObjectVector.vecLength = Integer.parseInt(header.substring(index + 10).trim());
}
while (indexInput.getFilePointer() < indexInput.length()) {
String term = indexInput.readString();
if (set.contains(term)) {
float[] v = new float[ObjectVector.vecLength];
for (int k = 0; k < v.length; k++) {
v[k] = Float.intBitsToFloat(indexInput.readInt());
}
map.put(term, v);
} else {
this.indexInput.seek(indexInput.getFilePointer() + ObjectVector.vecLength * 4);
}
}
Logger.getLogger(LuceneVectorStore.class.getName()).log(Level.INFO, "Prefetched {0} vectors", map.size());
return map;
}
/**
*
* @param term
* @return
* @throws IOException
*/
public float[] getFileVector(String term) throws IOException {
this.indexInput.seek(0);
String header = indexInput.readString(); //skip header
if ((header.equalsIgnoreCase("-dimensions"))) {
ObjectVector.vecLength = indexInput.readInt();
} else if (header.contains("-dimension")) {
int index = header.indexOf("-dimension");
ObjectVector.vecLength = Integer.parseInt(header.substring(index + 10).trim());
}
while (indexInput.getFilePointer() < indexInput.length()) {
String key = indexInput.readString();
if (key.equals(term)) {
float[] v = new float[ObjectVector.vecLength];
for (int k = 0; k < v.length; k++) {
v[k] = Float.intBitsToFloat(indexInput.readInt());
}
return v;
} else {
this.indexInput.seek(indexInput.getFilePointer() + ObjectVector.vecLength * 4);
}
}
throw new IOException("Vector for " + term + " not found");
}
/**
*
* @param vector
* @param n
* @return
* @throws IOException
*/
public List<SpaceResult> findFileSimilar(float[] vector, int n) throws IOException {
PriorityQueue<SpaceResult> queue = new PriorityQueue<>();
indexInput.seek(0);
String header = indexInput.readString(); //skip header
if ((header.equalsIgnoreCase("-dimensions"))) {
ObjectVector.vecLength = indexInput.readInt();
} else if (header.contains("-dimension")) {
int index = header.indexOf("-dimension");
ObjectVector.vecLength = Integer.parseInt(header.substring(index + 10).trim());
}
while (indexInput.getFilePointer() < indexInput.length()) {
String key = indexInput.readString();
float[] v = new float[ObjectVector.vecLength];
for (int k = 0; k < v.length; k++) {
v[k] = Float.intBitsToFloat(indexInput.readInt());
}
float score = VectorUtils.scalarProduct(vector, v);
if (queue.size() < n) {
queue.offer(new SpaceResult(key, score));
} else {
queue.poll();
queue.offer(new SpaceResult(key, score));
}
}
queue.poll();
List<SpaceResult> list = new ArrayList<>(queue);
Collections.sort(list);
return list;
}
/**
*
* @param term
* @return
*/
@Override
public float[] getVector(String term) {
return memory.get(term);
}
/**
*
* @param word
* @param n
* @return
*/
@Override
public List<SpaceResult> findSimilar(String word, int n) {
float[] v1 = memory.get(word);
if (v1 == null) {
Logger.getLogger(LuceneVectorStore.class.getName()).log(Level.WARNING, "No vector for term: {0}", word);
return new ArrayList<>();
}
PriorityQueue<SpaceResult> queue = new PriorityQueue<>();
Iterator<String> iterator = memory.keySet().iterator();
while (iterator.hasNext()) {
String key = iterator.next();
float[] v2 = memory.get(key);
float score = VectorUtils.scalarProduct(v1, v2);
if (queue.size() < n) {
queue.offer(new SpaceResult(key, score));
} else {
queue.poll();
queue.offer(new SpaceResult(key, score));
}
}
queue.poll();
List<SpaceResult> list = new ArrayList<>(queue);
Collections.sort(list);
return list;
}
/**
*
* @param vector
* @param n
* @return
*/
public List<SpaceResult> findSimilar(float[] vector, int n) {
PriorityQueue<SpaceResult> queue = new PriorityQueue<>();
Iterator<String> iterator = memory.keySet().iterator();
while (iterator.hasNext()) {
String key = iterator.next();
float[] v2 = memory.get(key);
float score = VectorUtils.scalarProduct(vector, v2);
if (queue.size() < n) {
queue.offer(new SpaceResult(key, score));
} else {
queue.poll();
queue.offer(new SpaceResult(key, score));
}
}
queue.poll();
List<SpaceResult> list = new ArrayList<>(queue);
Collections.sort(list);
return list;
}
/**
*
* @param map
* @param word
* @param n
* @return
*/
public List<SpaceResult> findSimilar(Map<String, float[]> map, String word, int n) {
float[] v1 = map.get(word);
if (v1 == null) {
Logger.getLogger(LuceneVectorStore.class.getName()).log(Level.WARNING, "No vector for term: {0}", word);
return new ArrayList<>();
}
PriorityQueue<SpaceResult> queue = new PriorityQueue<>();
Iterator<String> iterator = map.keySet().iterator();
while (iterator.hasNext()) {
String key = iterator.next();
float[] v2 = map.get(key);
float score = VectorUtils.scalarProduct(v1, v2);
if (queue.size() < n) {
queue.offer(new SpaceResult(key, score));
} else {
queue.poll();
queue.offer(new SpaceResult(key, score));
}
}
queue.poll();
List<SpaceResult> list = new ArrayList<>(queue);
Collections.sort(list);
return list;
}
/**
*
* @param map
* @param vector
* @param n
* @return
*/
public List<SpaceResult> findSimilar(Map<String, float[]> map, float[] vector, int n) {
PriorityQueue<SpaceResult> queue = new PriorityQueue<>();
Iterator<String> iterator = map.keySet().iterator();
while (iterator.hasNext()) {
String key = iterator.next();
float[] v2 = map.get(key);
float score = VectorUtils.scalarProduct(vector, v2);
if (queue.size() < n) {
queue.offer(new SpaceResult(key, score));
} else {
queue.poll();
queue.offer(new SpaceResult(key, score));
}
}
queue.poll();
List<SpaceResult> list = new ArrayList<>(queue);
Collections.sort(list);
return list;
}
}
