public void TestNameAtEnd() { var sentence = new[] { "My", "name", "is", "Anna" }; var sample = new NameSample(sentence, new[] { new Span(3, 4) }, false); Assert.AreEqual("My name is <START> Anna <END>", sample.ToString()); }
public override void run(string[] args) { if (args.Length == 0) { Console.WriteLine(Help); } else { NameFinderME[] nameFinders = new NameFinderME[args.Length]; for (int i = 0; i < nameFinders.Length; i++) { TokenNameFinderModel model = (new TokenNameFinderModelLoader()).load(new File(args[i])); nameFinders[i] = new NameFinderME(model); } ObjectStream <string> untokenizedLineStream = new PlainTextByLineStream(new InputStreamReader(Console.OpenStandardInput)); PerformanceMonitor perfMon = new PerformanceMonitor(System.err, "sent"); perfMon.start(); try { string line; while ((line = untokenizedLineStream.read()) != null) { string[] whitespaceTokenizerLine = WhitespaceTokenizer.INSTANCE.tokenize(line); // A new line indicates a new document, // adaptive data must be cleared for a new document if (whitespaceTokenizerLine.Length == 0) { foreach (NameFinderME nameFinder in nameFinders) { nameFinder.clearAdaptiveData(); } } IList <Span> names = new List <Span>(); foreach (TokenNameFinder nameFinder in nameFinders) { Collections.addAll(names, nameFinder.find(whitespaceTokenizerLine)); } // Simple way to drop intersecting spans, otherwise the // NameSample is invalid Span[] reducedNames = NameFinderME.dropOverlappingSpans(names.ToArray()); NameSample nameSample = new NameSample(whitespaceTokenizerLine, reducedNames, false); Console.WriteLine(nameSample.ToString()); perfMon.incrementCounter(); } } catch (IOException e) { CmdLineUtil.handleStdinIoError(e); } perfMon.stopAndPrintFinalResult(); } }