static Dictionary loadDict(File f) throws IOException { Dictionary dict = null; if (f != null) { CmdLineUtil.checkInputFile("abb dict", f); dict = new Dictionary(new FileInputStream(f)); } return dict; }
static Dictionary loadDict(File f) throws IOException { Dictionary dict = null; if (f != null) { CmdLineUtil.checkInputFile("abb dict", f); dict = new Dictionary(new FileInputStream(f)); } return dict; }
public T load(File modelFile) { long beginModelLoadingTime = System.currentTimeMillis(); CmdLineUtil.checkInputFile(modelName + " model", modelFile); System.err.print("Loading " + modelName + " model ... "); T model; try (InputStream modelIn = new BufferedInputStream( CmdLineUtil.openInFile(modelFile), CmdLineUtil.IO_BUFFER_SIZE)) { model = loadModel(modelIn); } catch (InvalidFormatException e) { System.err.println("failed"); throw new TerminateToolException(-1, "Model has invalid format", e); } catch (IOException e) { System.err.println("failed"); throw new TerminateToolException(-1, "IO error while loading model file '" + modelFile + "'", e); } long modelLoadingDuration = System.currentTimeMillis() - beginModelLoadingTime; System.err.printf("done (%.3fs)\n", modelLoadingDuration / 1000d); return model; } }
public void run(String[] args) { Params params = validateAndParseParams(args, Params.class); File dictInFile = params.getInputFile(); File dictOutFile = params.getOutputFile(); Charset encoding = params.getEncoding(); CmdLineUtil.checkInputFile("dictionary input file", dictInFile); CmdLineUtil.checkOutputFile("dictionary output file", dictOutFile); try (InputStreamReader in = new InputStreamReader(new FileInputStream(dictInFile), encoding); OutputStream out = new FileOutputStream(dictOutFile)) { Dictionary dict = Dictionary.parseOneEntryPerLine(in); dict.serialize(out); } catch (IOException e) { throw new TerminateToolException(-1, "IO error while reading training data or indexing data: " + e.getMessage(), e); } } }
public static TrainingParameters loadTrainingParameters(String paramFile, boolean supportSequenceTraining) { TrainingParameters params = null; if (paramFile != null) { checkInputFile("Training Parameter", new File(paramFile)); try (InputStream paramsIn = new FileInputStream(new File(paramFile))) { params = new opennlp.tools.util.TrainingParameters(paramsIn); } catch (IOException e) { throw new TerminateToolException(-1, "Error during parameters loading: " + e.getMessage(), e); } if (!TrainerFactory.isValid(params)) { throw new TerminateToolException(1, "Training parameters file '" + paramFile + "' is invalid!"); } TrainerFactory.TrainerType trainerType = TrainerFactory.getTrainerType(params); if (!supportSequenceTraining && trainerType.equals(TrainerFactory.TrainerType.EVENT_MODEL_SEQUENCE_TRAINER)) { throw new TerminateToolException(1, "Sequence training is not supported!"); } } return params; } }
public void run(String[] args) { Params params = validateAndParseParams(args, Params.class); File dictInFile = params.getInputFile(); CmdLineUtil.checkInputFile("dictionary input file", dictInFile); Path metadataPath = DictionaryMetadata.getExpectedMetadataLocation(dictInFile.toPath()); CmdLineUtil.checkInputFile("dictionary metadata (.info) input file", metadataPath.toFile()); MorfologikDictionayBuilder builder = new MorfologikDictionayBuilder(); try { builder.build(dictInFile.toPath(), params.getOverwrite(), params.getValidate(), params.getAcceptBOM(), params.getAcceptCR(), params.getIgnoreEmpty()); } catch (Exception e) { throw new TerminateToolException(-1, "Error while creating Morfologik POS Dictionay: " + e.getMessage(), e); } } }
@Override public ObjectStream<SentenceSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); CmdLineUtil.checkInputFile("Text", params.getTextFile()); NKJPSegmentationDocument segDoc = null; NKJPTextDocument textDoc = null; try { segDoc = NKJPSegmentationDocument.parse(params.getData()); textDoc = NKJPTextDocument.parse(params.getTextFile()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new NKJPSentenceSampleStream(segDoc, textDoc); } }
public ObjectStream<TokenSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); IrishSentenceBankDocument isbDoc = null; try { isbDoc = IrishSentenceBankDocument.parse(params.getData()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new IrishSentenceBankTokenSampleStream(isbDoc); } }
@Override public ObjectStream<SentenceSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); IrishSentenceBankDocument isbDoc = null; try { isbDoc = IrishSentenceBankDocument.parse(params.getData()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new IrishSentenceBankSentenceStream(isbDoc); } }
public ObjectStream<LanguageSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new LanguageDetectorSampleStream(lineStream); } }
public ObjectStream<ChunkSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new ChunkSampleStream(lineStream); } }
public ObjectStream<LemmaSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new LemmaSampleStream(lineStream); } }
public ObjectStream<NameSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new NameSampleDataStream(lineStream); } }
public ObjectStream<POSSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new WordTagSampleStream(lineStream); } }
public ObjectStream<Parse> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new ParseSampleStream(lineStream); } }
public ObjectStream<DocumentSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new DocumentSampleStream(lineStream); } }
public ObjectStream<SentenceSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new SentenceSampleStream(lineStream); } }
public ObjectStream<SentenceSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new MosesSentenceSampleStream(lineStream); } }
public ObjectStream<TokenSample> create(String[] args) { Parameters params = ArgumentParser.parse(args, Parameters.class); CmdLineUtil.checkInputFile("Data", params.getData()); InputStreamFactory sampleDataIn = CmdLineUtil.createInputStreamFactory(params.getData()); ObjectStream<String> lineStream = null; try { lineStream = new PlainTextByLineStream(sampleDataIn, params.getEncoding()); } catch (IOException ex) { CmdLineUtil.handleCreateObjectStreamError(ex); } return new TokenSampleStream(lineStream); } }
CmdLineUtil.checkInputFile("Data", params.getData());