//Convert CRFSharp output format to string list
private List <string> ConvertCRFTermOutToStringList(List <List <string> > inbuf, crf_seg_out[] crf_out)
{
StringBuilder sb = new StringBuilder();
for (int i = 0; i < inbuf.Count; i++)
{
sb.Append(inbuf[i][0]);
}
string strText = sb.ToString();
List <string> rstList = new List <string>();
for (int i = 0; i < crf_out.Length; i++)
{
if (crf_out[i] == null)
{
//No more result
break;
}
sb.Clear();
crf_seg_out crf_term_out = crf_out[i];
for (int j = 0; j < crf_term_out.Count; j++)
{
string str = strText.Substring(crf_term_out.tokenList[j].offset, crf_term_out.tokenList[j].length);
string strNE = crf_term_out.tokenList[j].strTag;
sb.Append(str);
if (strNE.Length > 0)
{
sb.Append("[" + strNE + "]");
}
sb.Append(" ");
}
rstList.Add(sb.ToString().Trim());
}
return(rstList);
}
private List <Token> LabelString(Instance instance, List <string> termList)
{
//Extract features from given text
List <List <string> > sinbuf = featureGenerator.GenerateFeature(termList);
//Call CRFSharp to predict word formation tags
int ret = crf.Segment(instance.crf_seg_out, instance.crf_tag, sinbuf);
//Only use 1st-best result
crf_seg_out item = instance.crf_seg_out[0];
if (ret < 0 || item.Count != termList.Count)
{
//CRF parsing is failed
string strMessage = "Failed to parse word formation by model. RetVal: " + ret.ToString() + ", Parsed Token Count: " + item.Count.ToString() + ", Input Token Count: " + termList.Count.ToString();
Console.WriteLine(strMessage);
return(null);
}
//Fill the token list
List <Token> tknList = new List <Token>();
for (int j = 0; j < item.Count; j++)
{
int offset = item.tokenList[j].offset;
int len = item.tokenList[j].length;
string strNE = item.tokenList[j].strTag;
Token token = new Token();
token.strTerm = termList[j];
token.rankId = int.Parse(strNE.Substring(strRankTagPrefix.Length));
tknList.Add(token);
}
return(tknList);
}
bool Decode(CRFSharpWrapper.DecoderArgs options)
{
var parallelOption = new ParallelOptions();
var watch = Stopwatch.StartNew();
if (File.Exists(options.strInputFileName) == false)
{
Logger.WriteLine("FAILED: Open {0} file failed.", options.strInputFileName);
return(false);
}
if (File.Exists(options.strModelFileName) == false)
{
Logger.WriteLine("FAILED: Open {0} file failed.", options.strModelFileName);
return(false);
}
var sr = new StreamReader(options.strInputFileName);
StreamWriter sw = null, swSeg = null;
if (options.strOutputFileName != null && options.strOutputFileName.Length > 0)
{
sw = new StreamWriter(options.strOutputFileName);
}
if (options.strOutputSegFileName != null && options.strOutputSegFileName.Length > 0)
{
swSeg = new StreamWriter(options.strOutputSegFileName);
}
//Create CRFSharp wrapper instance. It's a global instance
var crfWrapper = new CRFSharpWrapper.Decoder();
//Load encoded model from file
Logger.WriteLine("Loading model from {0}", options.strModelFileName);
crfWrapper.LoadModel(options.strModelFileName);
var queueRecords = new ConcurrentQueue <List <List <string> > >();
var queueSegRecords = new ConcurrentQueue <List <List <string> > >();
parallelOption.MaxDegreeOfParallelism = options.thread;
Parallel.For(0, options.thread, parallelOption, t =>
{
//Create decoder tagger instance. If the running environment is multi-threads, each thread needs a separated instance
var tagger = crfWrapper.CreateTagger(options.nBest, options.maxword);
tagger.set_vlevel(options.probLevel);
//Initialize result
var crf_out = new crf_seg_out[options.nBest];
for (var i = 0; i < options.nBest; i++)
{
crf_out[i] = new crf_seg_out(tagger.crf_max_word_num);
}
var 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)
{
var rstList = ConvertCRFTermOutToStringList(inbuf, crf_out);
while (peek != inbuf)
{
queueSegRecords.TryPeek(out peek);
}
for (int index = 0; index < rstList.Count; index++)
{
var item = rstList[index];
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);
}
OutputRawResultToFile(inbuf, crf_out, tagger, sw);
queueRecords.TryDequeue(out peek);
}
}
});
sr.Close();
if (sw != null)
{
sw.Close();
}
if (swSeg != null)
{
swSeg.Close();
}
watch.Stop();
Logger.WriteLine("Elapsed: {0} ms", watch.ElapsedMilliseconds);
return(true);
}
bool Decode(CRFSharpWrapper.DecoderArgs options)
{
ParallelOptions parallelOption = new ParallelOptions();
if (File.Exists(options.strInputFileName) == false)
{
Console.WriteLine("FAILED: Open {0} file failed.", options.strInputFileName);
return(false);
}
if (File.Exists(options.strModelFileName) == false)
{
Console.WriteLine("FAILED: Open {0} file failed.", options.strModelFileName);
return(false);
}
StreamReader sr = new StreamReader(options.strInputFileName);
StreamWriter sw = null, swSeg = null;
if (options.strOutputFileName != null && options.strOutputFileName.Length > 0)
{
sw = new StreamWriter(options.strOutputFileName);
}
if (options.strOutputSegFileName != null && options.strOutputSegFileName.Length > 0)
{
swSeg = new StreamWriter(options.strOutputSegFileName);
}
//Create CRFSharp wrapper instance. It's a global instance
CRFSharpWrapper.Decoder crfWrapper = new CRFSharpWrapper.Decoder();
//Load model from file
if (crfWrapper.LoadModel(options.strModelFileName) == false)
{
return(false);
}
ConcurrentQueue <List <List <string> > > queueRecords = new ConcurrentQueue <List <List <string> > >();
ConcurrentQueue <List <List <string> > > queueSegRecords = new ConcurrentQueue <List <List <string> > >();
parallelOption.MaxDegreeOfParallelism = options.thread;
Parallel.For(0, options.thread, parallelOption, t =>
{
//Create decoder tagger instance. If the running environment is multi-threads, each thread needs a separated instance
SegDecoderTagger tagger = crfWrapper.CreateTagger(options.nBest);
tagger.set_vlevel(options.probLevel);
//Initialize result
crf_seg_out[] crf_out = new crf_seg_out[options.nBest];
for (int i = 0; i < options.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);
}
OutputRawResultToFile(inbuf, crf_out, tagger, sw);
queueRecords.TryDequeue(out peek);
}
}
});
sr.Close();
if (sw != null)
{
sw.Close();
}
if (swSeg != null)
{
swSeg.Close();
}
return(true);
}
