@Before public void setup() throws IOException { InputStreamFactory in = new ResourceAsStreamFactory( ADParagraphStreamTest.class, "/opennlp/tools/formats/ad.sample"); try (ADChunkSampleStream stream = new ADChunkSampleStream(new PlainTextByLineStream(in, "UTF-8"))) { ChunkSample sample; while ((sample = stream.read()) != null) { samples.add(sample); } } }
private void chunkerCrossEval(TrainingParameters params, double expectedScore) throws IOException { ADChunkSampleStream samples = new ADChunkSampleStream(getLineSample(BOSQUE)); ChunkerCrossValidator cv = new ChunkerCrossValidator(LANG, params, new ChunkerFactory()); cv.evaluate(samples, 10); Assert.assertEquals(expectedScore, cv.getFMeasure().getFMeasure(), 0.0001d); }
public ObjectStream<ChunkSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); language = params.getLang(); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } ADChunkSampleStream sampleStream = new ADChunkSampleStream(lineStream); if (params.getStart() != null && params.getStart() > -1) { sampleStream.setStart(params.getStart()); } if (params.getEnd() != null && params.getEnd() > -1) { sampleStream.setEnd(params.getEnd()); } return sampleStream; } }
public ObjectStream<ChunkSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); language = params.getLang(); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } ADChunkSampleStream sampleStream = new ADChunkSampleStream(lineStream); if (params.getStart() != null && params.getStart() > -1) { sampleStream.setStart(params.getStart()); } if (params.getEnd() != null && params.getEnd() > -1) { sampleStream.setEnd(params.getEnd()); } return sampleStream; } }
public ObjectStream<ChunkSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); language = params.getLang(); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } ADChunkSampleStream sampleStream = new ADChunkSampleStream(lineStream); if (params.getStart() != null && params.getStart() > -1) { sampleStream.setStart(params.getStart()); } if (params.getEnd() != null && params.getEnd() > -1) { sampleStream.setEnd(params.getEnd()); } return sampleStream; } }