private void loadData(String filePath, final int[] inFactors, final int[] outFactors)
throws IOException {
FactorCollection factorCollection = FactorCollection.instance();
final String factorDelimiter = StaticData.instance().getFactorDelimiter();
VERBOSE.v(2, "Loading global lexical model from file " + filePath + "\n");
// m_inputFactors = new FactorMask(inFactors);
// m_outputFactors = new FactorMask(outFactors);
BufferedReader inFile = new BufferedReader(new FileReader(filePath));
// reading in data one line at a time
int lineNum = 0;
String line;
while ((line = inFile.readLine()) != null) {
++lineNum;
String token[] = Util.tokenize(line, " ");
if (token.length != 3) // format checking
{
StringBuilder errorMessage = new StringBuilder();
errorMessage
.append("Syntax error at ")
.append(filePath)
.append(":")
.append(lineNum)
.append('\n')
.append(line)
.append('\n');
UserMessage.add(errorMessage.toString());
System.exit(0);
}
// create the output word
Word outWord = new Word();
String[] factorString = Util.tokenize(token[0], factorDelimiter);
for (int i = 0; i < outFactors.length; i++) {
final FactorDirection direction = FactorDirection.Output;
final int factorType = outFactors[i];
final Factor factor = factorCollection.addFactor(direction, factorType, factorString[i]);
outWord.setFactor(factorType, factor);
}
// create the input word
Word inWord = new Word();
factorString = Util.tokenize(token[1], factorDelimiter);
for (int i = 0; i < inFactors.length; i++) {
final FactorDirection direction = FactorDirection.Input;
final int factorType = inFactors[i];
final Factor factor = factorCollection.addFactor(direction, factorType, factorString[i]);
inWord.setFactor(factorType, factor);
}
// maximum entropy feature score
float score = Float.valueOf(token[2]);
// std::cerr << "storing word " << *outWord << " " << *inWord <<
// " " << score << endl;
// store feature in hash
Map<Word, Float> dh = m_hash.get(outWord);
if (dh == null) {
dh = new HashMap<Word, Float>();
dh.put(inWord, score);
m_hash.put(outWord, dh);
} else {
dh.put(inWord, score);
outWord = null;
}
}
}
