public Alignment alignSentencePair(SentencePair sentencePair) {
Alignment alignment = new Alignment();
List<String> frenchWords = sentencePair.getFrenchWords();
List<String> englishWords = sentencePair.getEnglishWords();
int numFrenchWords = frenchWords.size();
int numEnglishWords = englishWords.size();
// Model 1 assumes all alignments are equally likely
// So we can just take the argMax of t(f|e) to get the englishMaxPosition
for (int frenchPosition = 0; frenchPosition < numFrenchWords; frenchPosition++) {
String f = frenchWords.get(frenchPosition);
int englishMaxPosition = -1;
double maxTranslationProb = translationProbs.getCount(f, NULL);
for (int englishPosition = 0; englishPosition < numEnglishWords; englishPosition++) {
String e = englishWords.get(englishPosition);
double translationProb = translationProbs.getCount(f, e);
if (translationProb > maxTranslationProb) {
maxTranslationProb = translationProb;
englishMaxPosition = englishPosition;
}
}
alignment.addAlignment(englishMaxPosition, frenchPosition, true);
}
return alignment;
}
/**
* Builds a Trellis over a sentence, by starting at the state State, and advancing through all
* legal extensions of each state already in the trellis. You should not have to modify this
* code (or even read it, really).
*/
private Trellis<State> buildTrellis(List<String> sentence) {
Trellis<State> trellis = new Trellis<State>();
trellis.setStartState(State.getStartState());
State stopState = State.getStopState(sentence.size() + 2);
trellis.setStopState(stopState);
Set<State> states = Collections.singleton(State.getStartState());
for (int position = 0; position <= sentence.size() + 1; position++) {
Set<State> nextStates = new HashSet<State>();
for (State state : states) {
if (state.equals(stopState)) continue;
LocalTrigramContext localTrigramContext =
new LocalTrigramContext(
sentence, position, state.getPreviousPreviousTag(), state.getPreviousTag());
Counter<String> tagScores = localTrigramScorer.getLogScoreCounter(localTrigramContext);
for (String tag : tagScores.keySet()) {
double score = tagScores.getCount(tag);
State nextState = state.getNextState(tag);
trellis.setTransitionCount(state, nextState, score);
nextStates.add(nextState);
}
}
// System.out.println("States: "+nextStates);
states = nextStates;
}
return trellis;
}
public Alignment alignSentencePair(SentencePair sentencePair) {
Alignment alignment = new Alignment();
List<String> frenchWords = sentencePair.getFrenchWords();
List<String> englishWords = sentencePair.getEnglishWords();
int numFrenchWords = frenchWords.size();
int numEnglishWords = englishWords.size();
for (int frenchPosition = 0; frenchPosition < numFrenchWords; frenchPosition++) {
String f = frenchWords.get(frenchPosition);
int englishMaxPosition = frenchPosition;
if (englishMaxPosition >= numEnglishWords)
englishMaxPosition = -1; // map French word to BASELINE if c(f,e) = 0 for all English words
double maxDice = 0;
for (int englishPosition = 0; englishPosition < numEnglishWords; englishPosition++) {
String e = englishWords.get(englishPosition);
double dice = getDiceCoefficient(f,e);
if (dice > maxDice) {
maxDice = dice;
englishMaxPosition = englishPosition;
}
}
alignment.addAlignment(englishMaxPosition, frenchPosition, true);
}
return alignment;
}
public Alignment alignSentencePair(SentencePair sentencePair) {
Alignment alignment = new Alignment();
List<String> frenchWords = sentencePair.getFrenchWords();
List<String> englishWords = sentencePair.getEnglishWords();
int numFrenchWords = frenchWords.size();
int numEnglishWords = englishWords.size();
for (int frenchPosition = 0; frenchPosition < numFrenchWords; frenchPosition++) {
String f = frenchWords.get(frenchPosition);
int englishMaxPosition = frenchPosition;
if (englishMaxPosition >= numEnglishWords)
englishMaxPosition = -1; // map French word to BASELINE if c(f,e) = 0 for all English words
double maxConditionalProb = 0;
for (int englishPosition = 0; englishPosition < numEnglishWords; englishPosition++) {
String e = englishWords.get(englishPosition);
double conditionalGivenEnglish = collocationCounts.getCount(f, e) / (eCounts.getCount(e));
if (conditionalGivenEnglish > maxConditionalProb) {
maxConditionalProb = conditionalGivenEnglish;
englishMaxPosition = englishPosition;
}
}
alignment.addAlignment(englishMaxPosition, frenchPosition, true);
}
return alignment;
}
public static List<String> toTagList(List<State> states) {
List<String> tags = new ArrayList<String>();
if (states.size() > 0) {
tags.add(states.get(0).getPreviousPreviousTag());
for (State state : states) {
tags.add(state.getPreviousTag());
}
}
return tags;
}
private static List<SentencePair> readSentencePairs(String path, int maxSentencePairs) {
List<SentencePair> sentencePairs = new ArrayList<SentencePair>();
List<String> baseFileNames = getBaseFileNames(path);
for (String baseFileName : baseFileNames) {
if (sentencePairs.size() >= maxSentencePairs)
continue;
sentencePairs.addAll(readSentencePairs(baseFileName));
}
return sentencePairs;
}
public String toString() {
StringBuilder sb = new StringBuilder();
for (int position = 0; position < words.size(); position++) {
String word = words.get(position);
String tag = tags.get(position);
sb.append(word);
sb.append("_");
sb.append(tag);
}
return sb.toString();
}
public String toString() {
StringBuilder sb = new StringBuilder();
for (int englishPosition = 0; englishPosition < englishWords.size(); englishPosition++) {
String englishWord = englishWords.get(englishPosition);
sb.append(englishPosition);
sb.append(":");
sb.append(englishWord);
sb.append(" ");
}
sb.append("\n");
for (int frenchPosition = 0; frenchPosition < frenchWords.size(); frenchPosition++) {
String frenchWord = frenchWords.get(frenchPosition);
sb.append(frenchPosition);
sb.append(":");
sb.append(frenchWord);
sb.append(" ");
}
sb.append("\n");
return sb.toString();
}
// pretty-print a pair of taggings for a sentence, possibly suppressing the tags which correctly
// match
private static String alignedTaggings(
List<String> words,
List<String> goldTags,
List<String> guessedTags,
boolean suppressCorrectTags) {
StringBuilder goldSB = new StringBuilder("Gold Tags: ");
StringBuilder guessedSB = new StringBuilder("Guessed Tags: ");
StringBuilder wordSB = new StringBuilder("Words: ");
for (int position = 0; position < words.size(); position++) {
equalizeLengths(wordSB, goldSB, guessedSB);
String word = words.get(position);
String gold = goldTags.get(position);
String guessed = guessedTags.get(position);
wordSB.append(word);
if (position < words.size() - 1) wordSB.append(' ');
boolean correct = (gold.equals(guessed));
if (correct && suppressCorrectTags) continue;
guessedSB.append(guessed);
goldSB.append(gold);
}
return goldSB + "\n" + guessedSB + "\n" + wordSB;
}
private static void labelTestSet(
POSTagger posTagger, List<TaggedSentence> testSentences, String path) throws Exception {
BufferedWriter writer = new BufferedWriter(new FileWriter(path));
for (TaggedSentence sentence : testSentences) {
List<String> words = sentence.getWords();
List<String> guessedTags = posTagger.tag(words);
for (int i = 0; i < words.size(); i++) {
writer.write(words.get(i) + "\t" + guessedTags.get(i) + "\n");
}
writer.write("\n");
}
writer.close();
}
private static void evaluateTagger(
POSTagger posTagger,
List<TaggedSentence> taggedSentences,
Set<String> trainingVocabulary,
boolean verbose) {
double numTags = 0.0;
double numTagsCorrect = 0.0;
double numUnknownWords = 0.0;
double numUnknownWordsCorrect = 0.0;
int numDecodingInversions = 0;
for (TaggedSentence taggedSentence : taggedSentences) {
List<String> words = taggedSentence.getWords();
List<String> goldTags = taggedSentence.getTags();
List<String> guessedTags = posTagger.tag(words);
for (int position = 0; position < words.size() - 1; position++) {
String word = words.get(position);
String goldTag = goldTags.get(position);
String guessedTag = guessedTags.get(position);
if (guessedTag.equals(goldTag)) numTagsCorrect += 1.0;
numTags += 1.0;
if (!trainingVocabulary.contains(word)) {
if (guessedTag.equals(goldTag)) numUnknownWordsCorrect += 1.0;
numUnknownWords += 1.0;
}
}
double scoreOfGoldTagging = posTagger.scoreTagging(taggedSentence);
double scoreOfGuessedTagging = posTagger.scoreTagging(new TaggedSentence(words, guessedTags));
if (scoreOfGoldTagging > scoreOfGuessedTagging) {
numDecodingInversions++;
if (verbose)
System.out.println(
"WARNING: Decoder suboptimality detected. Gold tagging has higher score than guessed tagging.");
}
if (verbose) System.out.println(alignedTaggings(words, goldTags, guessedTags, true) + "\n");
}
System.out.println(
"Tag Accuracy: "
+ (numTagsCorrect / numTags)
+ " (Unknown Accuracy: "
+ (numUnknownWordsCorrect / numUnknownWords)
+ ") Decoder Suboptimalities Detected: "
+ numDecodingInversions);
}
private static List<TaggedSentence> readTaggedSentences(String path, boolean hasTags)
throws Exception {
List<TaggedSentence> taggedSentences = new ArrayList<TaggedSentence>();
BufferedReader reader = new BufferedReader(new FileReader(path));
String line = "";
List<String> words = new LinkedList<String>();
List<String> tags = new LinkedList<String>();
while ((line = reader.readLine()) != null) {
if (line.equals("")) {
taggedSentences.add(
new TaggedSentence(
new BoundedList<String>(words, START_WORD, STOP_WORD),
new BoundedList<String>(tags, START_WORD, STOP_WORD)));
words = new LinkedList<String>();
tags = new LinkedList<String>();
} else {
String[] fields = line.split("\\s+");
words.add(fields[0]);
tags.add(hasTags ? fields[1] : "");
}
}
System.out.println("Read " + taggedSentences.size() + " sentences.");
return taggedSentences;
}
private List<String> stripBoundaryTags(List<String> tags) {
return tags.subList(2, tags.size() - 2);
}
public int size() {
return words.size();
}
private CounterMap<String,String> trainEM(int maxIterations) {
Set<String> englishVocab = new HashSet<String>();
Set<String> frenchVocab = new HashSet<String>();
CounterMap<String,String> translations = new CounterMap<String,String>();
englishVocab.add(NULL);
int iteration = 0;
final double thresholdProb = 0.0001;
for (SentencePair sentencePair : trainingSentencePairs) {
List<String> frenchWords = sentencePair.getFrenchWords();
List<String> englishWords = sentencePair.getEnglishWords();
// add words from list to vocabulary sets
englishVocab.addAll(englishWords);
frenchVocab.addAll(frenchWords);
}
System.out.println("Ready");
// We need to initialize translations.getCount(f,e) uniformly
// t(f|e) summed over all e in {E + NULL} = 1
final double initialCount = 1.0 / englishVocab.size();
while(iteration < maxIterations) {
CounterMap<String,String> counts = new CounterMap<String,String>(); // set count(f|e) to 0 for all e,f
Counter<String> totalEnglish = new Counter<String>(); // set total(e) to 0 for all e
// E-step: loop over all sentences and update counts
for (SentencePair sentencePair : trainingSentencePairs) {
List<String> frenchWords = sentencePair.getFrenchWords();
List<String> englishWords = sentencePair.getEnglishWords();
int numFrenchWords = frenchWords.size();
int numEnglishWords = englishWords.size();
Counter<String> sTotalF = new Counter<String>();
// compute normalization constant sTotalF
for (int frenchPosition = 0; frenchPosition < numFrenchWords; frenchPosition++) {
String f = frenchWords.get(frenchPosition);
// initialize and compute for English = NULL
if (!translations.containsKey(f) && initialize)
translations.setCount(f, NULL, initialCount);
else if (!translations.containsKey(f))
translations.setCount(f, NULL, thresholdProb);
sTotalF.incrementCount(f, translations.getCount(f, NULL));
for (int englishPosition = 0; englishPosition < numEnglishWords; englishPosition++) {
String e = englishWords.get(englishPosition);
if (!(translations.getCounter(f)).containsKey(e) && initialize)
translations.setCount(f, e, initialCount);
else if (!(translations.getCounter(f)).containsKey(e))
translations.setCount(f, e, thresholdProb);
sTotalF.incrementCount(f, translations.getCount(f, e));
}
}
// collect counts in counts and totalEnglish
for (int frenchPosition = 0; frenchPosition < numFrenchWords; frenchPosition++) {
String f = frenchWords.get(frenchPosition);
// collect counts for English = NULL
double count = translations.getCount(f, NULL) / sTotalF.getCount(f);
counts.incrementCount(NULL, f, count);
totalEnglish.incrementCount(NULL, count);
for (int englishPosition = 0; englishPosition < numEnglishWords; englishPosition++) {
String e = englishWords.get(englishPosition);
count = translations.getCount(f, e) / sTotalF.getCount(f);
counts.incrementCount(e, f, count);
totalEnglish.incrementCount(e, count);
}
}
} // end of E-step
System.out.println("Completed E-step");
// M-step: update probabilities with counts from E-step and check for convergence
iteration++;
for (String e : counts.keySet()) {//englishVocab) {
double normalizer = totalEnglish.getCount(e);
for (String f : (counts.getCounter(e)).keySet()) {//frenchVocab) {
// To speed implementation, we want to update translations only when count / normalizer > threshold
double prob = counts.getCount(e, f) / normalizer;
if (!initialize) {
if (prob > thresholdProb)
translations.setCount(f, e, prob);
else
(translations.getCounter(f)).removeKey(e);
}
else {
translations.setCount(f, e, prob);
}
}
}
System.out.println("Completed iteration " + iteration);
} // end of M-step
System.out.println("Trained!");
return translations;
}