public SegDecoderTagger CreateTagger()
{
if (modelReader == null)
{
return(null);
}
SegDecoderTagger tagger = new SegDecoderTagger();
tagger.init_by_model(modelReader);
return(tagger);
}
//Segment given text
public int Segment(crf_seg_out[] pout, //segment result
SegDecoderTagger tagger, //Tagger per thread
List <List <string> > inbuf //feature set for segment
)
{
int ret = 0;
if (inbuf.Count == 0)
{
//Empty input string
return(Utils.ERROR_SUCCESS);
}
ret = tagger.reset();
if (ret < 0)
{
return(ret);
}
ret = tagger.add(inbuf);
if (ret < 0)
{
return(ret);
}
//parse
ret = tagger.parse();
if (ret < 0)
{
return(ret);
}
//wrap result
ret = tagger.output(pout);
if (ret < 0)
{
return(ret);
}
return(Utils.ERROR_SUCCESS);
}
public string Predict(DecoderArgs _property)
{
ParallelOptions parallelOption = new ParallelOptions();
StringWriter sw = null, swSeg = null;
if (_property.outputstyle == 0)
{
sw = new StringWriter();
}
else
{
swSeg = new StringWriter();
}
ConcurrentQueue <List <List <string> > > queueRecords = new ConcurrentQueue <List <List <string> > >();
ConcurrentQueue <List <List <string> > > queueSegRecords = new ConcurrentQueue <List <List <string> > >();
StringReader sr = new StringReader(_property.predictstring);
parallelOption.MaxDegreeOfParallelism = _property.thread;
Parallel.For(0, parallelOption.MaxDegreeOfParallelism, t =>
{
//Create decoder tagger instance. If the running environment is multi-threads, each thread needs a separated instance
SegDecoderTagger tagger = _crfWrapper.CreateTagger();
tagger.set_nbest(_property.nbest);
tagger.set_vlevel(_property.probLevel);
//Initialize result
crf_seg_out[] crf_out = new crf_seg_out[_property.nbest];
for (int i = 0; i < _property.nbest; i++)
{
crf_out[i] = new crf_seg_out();
}
List <List <string> > inbuf = new List <List <string> >();
while (true)
{
lock (rdLocker)
{
if (ReadRecord(inbuf, sr) == false)
{
break;
}
queueRecords.Enqueue(inbuf);
queueSegRecords.Enqueue(inbuf);
}
//Call CRFSharp wrapper to predict given string's tags
if (swSeg != null)
{
_crfWrapper.Segment(crf_out, tagger, inbuf);
}
else
{
_crfWrapper.Segment((crf_term_out[])crf_out, (DecoderTagger)tagger, inbuf);
}
List <List <string> > peek = null;
//Save segmented tagged result into file
if (swSeg != null)
{
List <string> rstList = ConvertCRFTermOutToStringList(inbuf, crf_out);
while (peek != inbuf)
{
queueSegRecords.TryPeek(out peek);
}
foreach (string item in rstList)
{
swSeg.WriteLine(item);
}
queueSegRecords.TryDequeue(out peek);
peek = null;
}
//Save raw tagged result (with probability) into file
if (sw != null)
{
while (peek != inbuf)
{
queueRecords.TryPeek(out peek);
}
OutputRawResult(inbuf, crf_out, tagger, sw);
queueRecords.TryDequeue(out peek);
}
}
});
if (sw != null)
{
sw.Close();
return(sw.ToString());
}
if (swSeg != null)
{
swSeg.Close();
return(swSeg.ToString());
}
return("");
}
//Output raw result with probability
private string OutputRawResult(List <List <string> > inbuf, crf_term_out[] crf_out, SegDecoderTagger tagger, StringWriter sw)
{
//best N-result
for (int k = 0; k < crf_out.Length; k++)
{
if (crf_out[k] == null)
{
//No more result
break;
}
StringBuilder sb = new StringBuilder();
crf_term_out crf_seg_out = crf_out[k];
//Show the entire sequence probability
//For each token
for (int i = 0; i < inbuf.Count; i++)
{
//Show all features
for (int j = 0; j < inbuf[i].Count; j++)
{
sb.Append(inbuf[i][j]);
sb.Append("\t");
}
//Show the best result and its probability
sb.Append(crf_seg_out.result_[i]);
if (tagger.vlevel_ > 1)
{
sb.Append("\t");
sb.Append(crf_seg_out.weight_[i]);
//Show the probability of all tags
sb.Append("\t");
for (int j = 0; j < tagger.ysize_; j++)
{
sb.Append(tagger.yname(j));
sb.Append("/");
sb.Append(tagger.prob(i, j));
if (j < tagger.ysize_ - 1)
{
sb.Append("\t");
}
}
}
sb.AppendLine();
}
if (tagger.vlevel_ > 0)
{
sw.WriteLine("#{0}", crf_seg_out.prob);
}
sw.WriteLine(sb.ToString().Trim());
sw.WriteLine();
}
return(sw.ToString());
}
