本文整理汇总了Java中com.clearnlp.nlp.NLPGetter类的典型用法代码示例。如果您正苦于以下问题:Java NLPGetter类的具体用法?Java NLPGetter怎么用?Java NLPGetter使用的例子?那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。
NLPGetter类属于com.clearnlp.nlp包,在下文中一共展示了NLPGetter类的20个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于我们的系统推荐出更棒的Java代码示例。
示例1: setupComponents
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
private static AbstractComponent[] setupComponents() {
AbstractComponent[] components = null;
try {
String language = AbstractReader.LANG_EN;
String modelType = "general-en";
AbstractComponent tagger = NLPGetter.getComponent(modelType, language, NLPMode.MODE_POS);
AbstractComponent morphological = NLPGetter.getComponent(modelType, language, NLPMode.MODE_MORPH);
AbstractComponent parser = NLPGetter.getComponent(modelType, language, NLPMode.MODE_DEP);
AbstractComponent identifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_PRED);
AbstractComponent classifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_ROLE);
AbstractComponent labeler = NLPGetter.getComponent(modelType, language, NLPMode.MODE_SRL);
components = new AbstractComponent[]{tagger, morphological, parser, identifier, classifier, labeler};//, identifier, classifier, labeler};
} catch (Exception e) {
throw new RuntimeException(e);
}
return components;
}
开发者ID:marcusklang,项目名称:langforia,代码行数:21,代码来源:ClearNLP.java
示例2: call
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public String call()
{
AbstractSegmenter segmenter = NLPGetter.getSegmenter(s_language, NLPGetter.getTokenizer(s_language));
BufferedReader reader = new BufferedReader(new StringReader(s_line));
StringBuilder build = new StringBuilder();
for (List<String> tokens : segmenter.getSentences(reader))
{
if (tokens.size() < 3) continue;
build.append(UTArray.join(tokens, "\n"));
build.append("\n\n");
}
try
{
reader.close();
}
catch (IOException e) {e.printStackTrace();}
return build.toString();
}
开发者ID:clearnlp,项目名称:clearnlp,代码行数:22,代码来源:DemoMultiThread.java
示例3: decode
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public void decode(LineReader reader, PrintStream fout, AbstractTokenizer tokenizer, AbstractComponent[] components)
{
String sentence, mode = getMode();
DEPTree tree;
while ((sentence = reader.next()) != null)
{
if (sentence.trim().equals(UNConstant.EMPTY)) continue;
tree = NLPGetter.toDEPTree(tokenizer.getTokens(sentence));
for (AbstractComponent component : components)
component.process(tree);
fout.println(toString(tree, mode)+"\n");
}
}
开发者ID:clearnlp,项目名称:clearnlp,代码行数:17,代码来源:NLPDecoder.java
示例4: Tokenizer
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public Tokenizer(String[] args)
{
initArgs(args);
try
{
AbstractTokenizer tokenizer = NLPGetter.getTokenizer(s_language);
AbstractSegmenter segmenter = i_format.equals(AbstractReader.TYPE_RAW) ? NLPGetter.getSegmenter(s_language, tokenizer) : null;
List<String[]> filenames = getFilenames(s_inputPath, s_inputExt, s_outputExt);
boolean outLine = o_format.equals(AbstractReader.TYPE_LINE);
tokenizer.setTwit(b_twit);
for (String[] io : filenames)
{
System.out.println(io[0]);
tokenize(tokenizer, segmenter, io[0], io[1], outLine);
}
}
catch (IOException e) {e.printStackTrace();}
}
开发者ID:clearnlp,项目名称:clearnlp,代码行数:21,代码来源:Tokenizer.java
示例5: process
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public void process(AbstractTokenizer tokenizer, AbstractComponent[] components, BufferedReader reader, PrintStream fout)
{
AbstractSegmenter segmenter = NLPGetter.getSegmenter(language, tokenizer);
DEPTree tree;
for (List<String> tokens : segmenter.getSentences(reader))
{
tree = NLPGetter.toDEPTree(tokens);
for (AbstractComponent component : components)
component.process(tree);
fout.println(tree.toStringSRL()+"\n");
}
fout.close();
}
开发者ID:clearnlp,项目名称:clearnlp-demo,代码行数:18,代码来源:DemoNLPDecode.java
示例6: DemoCSenTypeClassifierEN
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public DemoCSenTypeClassifierEN(String inputFile, String outputFile) throws Exception
{
final String language = AbstractReader.LANG_EN;
final String modelType = "general-en";
// final String modelType = "medical-en";
AbstractSegmenter segmenter = NLPGetter.getSegmenter(language, NLPGetter.getTokenizer(language));
AbstractComponent tagger = NLPGetter.getComponent(modelType, language, NLPMode.MODE_POS);
AbstractComponent parser = NLPGetter.getComponent(modelType, language, NLPMode.MODE_DEP);
CSenTypeClassifierEN typer = new CSenTypeClassifierEN();
BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(inputFile)));
PrintStream fout = new PrintStream(new BufferedOutputStream(new FileOutputStream(outputFile)));
process(reader, fout, segmenter, tagger, parser, typer);
reader.close();
fout.close();
}
开发者ID:clearnlp,项目名称:clearnlp-demo,代码行数:20,代码来源:DemoCSenTypeClassifierEN.java
示例7: process
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public void process(BufferedReader reader, PrintStream fout, AbstractSegmenter segmenter, AbstractComponent tagger, AbstractComponent parser, CSenTypeClassifierEN typer)
{
DEPTree tree;
for (List<String> tokens : segmenter.getSentences(reader))
{
tree = NLPGetter.toDEPTree(tokens); // put tokens into dependency tree
tagger.process(tree); // part-of-speech tagging
parser.process(tree); // dependency parsing
tree.setDependents();
for (DEPNode root : tree.getRoots())
{
if (typer.isInterrogative(root))
{
fout.println(UTArray.join(tokens, " ")+"\n");
fout.println(tree.toStringDEP()+"\n");
break;
}
}
}
}
开发者ID:clearnlp,项目名称:clearnlp-demo,代码行数:23,代码来源:DemoCSenTypeClassifierEN.java
示例8: DemoMultiParse
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public DemoMultiParse(String modelType, String inputFile, String outputFile) throws Exception
{
AbstractTokenizer tokenizer = NLPGetter.getTokenizer(language);
AbstractComponent tagger = NLPGetter.getComponent(modelType, language, NLPMode.MODE_POS);
AbstractComponent parser = NLPGetter.getComponent(modelType, language, NLPMode.MODE_DEP);
AbstractComponent identifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_PRED);
AbstractComponent classifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_ROLE);
AbstractComponent labeler = NLPGetter.getComponent(modelType, language, NLPMode.MODE_SRL);
AbstractComponent[] preComponents = {tagger}; // components used before parsing
AbstractComponent[] postComponents = {identifier, classifier, labeler}; // components used after parsing
String sentence = "I know you know who I know.";
process(tokenizer, (AbstractDEPParser)parser, preComponents, postComponents, sentence);
process(tokenizer, (AbstractDEPParser)parser, preComponents, postComponents, UTInput.createBufferedFileReader(inputFile), UTOutput.createPrintBufferedFileStream(outputFile));
}
开发者ID:clearnlp,项目名称:clearnlp-demo,代码行数:17,代码来源:DemoMultiParse.java
示例9: loadPosTagger
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public void loadPosTagger() throws IOException {
String model_path = prop.getProperty("pos.model");
if (!model_path.toLowerCase().endsWith(".zip"))
tagger = (AbstractPOSTagger) NLPGetter.getComponent(model_path, AbstractReader.LANG_EN, NLPMode.MODE_POS);
else
tagger = (AbstractPOSTagger) NLPGetter.getComponent(new ZipFile(model_path), AbstractReader.LANG_EN, NLPMode.MODE_POS);
}
开发者ID:hakchul77,项目名称:irnlp_toolkit,代码行数:8,代码来源:ClearNlpWrapper.java
示例10: loadPredIdentifier
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
private void loadPredIdentifier() throws IOException {
String model_path = prop.getProperty("pred.model");
if (!model_path.toLowerCase().endsWith(".zip"))
identifier = (AbstractPredicateIdentifier) NLPGetter.getComponent(model_path, AbstractReader.LANG_EN, NLPMode.MODE_PRED);
else
identifier = (AbstractPredicateIdentifier) NLPGetter.getComponent(new ZipFile(model_path), AbstractReader.LANG_EN, NLPMode.MODE_PRED);
}
开发者ID:hakchul77,项目名称:irnlp_toolkit,代码行数:8,代码来源:ClearNlpWrapper.java
示例11: loadRoleClassifier
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
private void loadRoleClassifier() throws IOException {
String model_path = prop.getProperty("role.model");
if (!model_path.toLowerCase().endsWith(".zip"))
classifier = (AbstractRolesetClassifier) NLPGetter.getComponent(model_path, AbstractReader.LANG_EN, NLPMode.MODE_ROLE);
else
classifier = (AbstractRolesetClassifier) NLPGetter.getComponent(new ZipFile(model_path), AbstractReader.LANG_EN, NLPMode.MODE_ROLE);
}
开发者ID:hakchul77,项目名称:irnlp_toolkit,代码行数:8,代码来源:ClearNlpWrapper.java
示例12: loadSrlLabeler
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public void loadSrlLabeler() throws IOException {
loadPredIdentifier();
loadRoleClassifier();
String model_path = prop.getProperty("srl.model");
if (!model_path.toLowerCase().endsWith(".zip"))
labeler = (AbstractSRLabeler) NLPGetter.getComponent(model_path, AbstractReader.LANG_EN, NLPMode.MODE_SRL);
else
labeler = (AbstractSRLabeler) NLPGetter.getComponent(new ZipFile(model_path), AbstractReader.LANG_EN, NLPMode.MODE_SRL);
}
开发者ID:hakchul77,项目名称:irnlp_toolkit,代码行数:10,代码来源:ClearNlpWrapper.java
示例13: main
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
/**
* Main function
*
* @param args
* @throws Exception
*/
public static void main(String[] args) throws Exception {
// data input
String text = "Samsung Electronics is a South Korean multinational electronics company in Suwon, South Korea.";
// model loading
ClearNlpWrapper nlp = new ClearNlpWrapper(Env.CLEARNLP_CFG);
nlp.loadAll("tokenize, ssplit, pos, parse, srl");
// task run
for (List<String> toks : nlp.detect(text)) {
DEPTree units = NLPGetter.toDEPTree(toks);
units = nlp.tag(units);
units = nlp.parse(units);
units = nlp.label(units);
System.out.println("-toStringSRL--------------------------------------------------------------------");
System.out.println(units.toStringSRL());
System.out.println("-dependent Nodes----------------------------------------------------------------");
StringBuffer sb = new StringBuffer();
for (DEPNode verb : ClearNlpWrapper.getAllVerbs(units)) {
sb.append("->(Verb) " + ClearNlpWrapper.toTaggedWord(verb) + "\n");
for (SRLNode dep : ClearNlpWrapper.getDependents(verb)) {
Map<String, Object> m = dep.toMap();
sb.append(String.format(" ->(Node) id=%d, form=%s, pos=%s, drel=%s",
m.get("id"), m.get("form"), m.get("pos"), m.get("drel")));
if (m.get("srel") != null)
sb.append(String.format(", srel=%s, sfunc=%s", m.get("srel"), m.get("sfunc")));
sb.append("\n");
}
}
System.out.println(JString.trimAndIndent(sb.toString(), 2));
}
}
开发者ID:hakchul77,项目名称:irnlp_toolkit,代码行数:41,代码来源:SemanticRoleLabeling.java
示例14: testClearNlpWrapperForBasic
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
/**
* ClearNlpWrapper Test for basic functions
*
* @throws IOException
*/
public void testClearNlpWrapperForBasic() throws IOException {
System.out.println("\n----- testClearNlpWrapperForBasic() ------------------------------");
if (!TEST_BASIC)
return;
String text = "Samsung Electronics is a South Korean multinational electronics company in Suwon, South Korea.";
text += " It is the flagship subsidiary of the Samsung Group.";
ClearNlpWrapper nlp = new ClearNlpWrapper(Env.CLEARNLP_CFG);
nlp.loadAll("tokenize, ssplit, pos, parse");
assertTrue(nlp.tokenizer != null);
assertTrue(nlp.detector != null);
assertTrue(nlp.tagger != null);
assertTrue(nlp.parser != null);
assertEquals(2, nlp.detect(text).size());
for (List<String> words : nlp.detect(text)) {
DEPTree units = NLPGetter.toDEPTree(words);
System.out.println("\n[Sentence] " + JString.join(" ", words));
assertEquals(words.size(), ClearNlpWrapper.toTaggedWords(nlp.tag(units)).size());
System.out.println(" <Tagged> " + JString.join(" ", ClearNlpWrapper.toTaggedWords(nlp.tag(units))));
assertEquals(words.size(), ClearNlpWrapper.toTypedDependencies(nlp.parse(units)).size());
System.out.println(" <Parsed> " + JString.join("; ", ClearNlpWrapper.toTypedDependencies(nlp.parse(units))));
System.out.println("-parsed SemanticGraph-----------------------------------------------------------");
SemanticGraph sgraph = StanfordNlpWrapper.toSemanticGraph(ClearNlpWrapper.toTypedDependencies(nlp.parse(units)), false);
assertEquals(words.size(), JString.trimAndIndent(sgraph.toString(), 2).split("\n").length);
System.out.println(JString.trimAndIndent(sgraph.toString(), 2));
}
}
开发者ID:hakchul77,项目名称:irnlp_toolkit,代码行数:36,代码来源:TestClearNLP.java
示例15: getSegmenter
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
protected AbstractSegmenter getSegmenter(Element eConfig, boolean twit) throws IOException
{
AbstractTokenizer tokenizer = getTokenizer(eConfig, twit);
String language = getLanguage(eConfig);
return NLPGetter.getSegmenter(language, tokenizer);
}
开发者ID:clearnlp,项目名称:clearnlp,代码行数:8,代码来源:NLPDecoder.java
示例16: getTokenizer
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
protected AbstractTokenizer getTokenizer(Element eConfig, boolean twit) throws IOException
{
AbstractTokenizer tokenizer = NLPGetter.getTokenizer(getLanguage(eConfig));
tokenizer.setTwit(twit);
return tokenizer;
}
开发者ID:clearnlp,项目名称:clearnlp,代码行数:8,代码来源:NLPDecoder.java
示例17: convert
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public void convert(String headruleFile, String language, String mergeLabels, String inputPath, String parseExt, String propExt, String senseExt, String vclassExt, String nameExt, String outputExt) throws Exception
{
AbstractComponent morph = NLPGetter.getMPAnalyzer(s_language);
AbstractC2DConverter c2d = NLPGetter.getC2DConverter(s_language, s_headruleFile, s_mergeLabels);
convertRec(c2d, morph, language, inputPath, parseExt, propExt, senseExt, vclassExt, nameExt, outputExt);
}
开发者ID:clearnlp,项目名称:clearnlp,代码行数:8,代码来源:C2DConvertMulti.java
示例18: C2DConvert
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public C2DConvert(String[] args) throws Exception
{
initArgs(args);
AbstractC2DConverter c2d = NLPGetter.getC2DConverter(s_language, s_headruleFile, s_mergeLabels);
AbstractComponent morph = NLPGetter.getMPAnalyzer(s_language);
List<String[]> filenames = getFilenames(s_inputPath, s_inputExt, s_outputExt);
int n;
for (String[] io : filenames)
{
n = convert(c2d, morph, s_language, io[0], io[1], b_normalize);
System.out.printf("%s: %d trees\n", io[0], n);
}
}
开发者ID:clearnlp,项目名称:clearnlp,代码行数:16,代码来源:C2DConvert.java
示例19: DemoNLPDecode
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public DemoNLPDecode(String modelType, String inputFile, String outputFile) throws Exception
{
AbstractTokenizer tokenizer = NLPGetter.getTokenizer(language);
AbstractComponent tagger = NLPGetter.getComponent(modelType, language, NLPMode.MODE_POS);
AbstractComponent parser = NLPGetter.getComponent(modelType, language, NLPMode.MODE_DEP);
AbstractComponent identifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_PRED);
AbstractComponent classifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_ROLE);
AbstractComponent labeler = NLPGetter.getComponent(modelType, language, NLPMode.MODE_SRL);
AbstractComponent[] components = {tagger, parser, identifier, classifier, labeler};
String sentence = "I'd like to meet Dr. Choi.";
process(tokenizer, components, sentence);
process(tokenizer, components, UTInput.createBufferedFileReader(inputFile), UTOutput.createPrintBufferedFileStream(outputFile));
}
开发者ID:clearnlp,项目名称:clearnlp-demo,代码行数:16,代码来源:DemoNLPDecode.java
示例20: DemoMultiThread
import com.clearnlp.nlp.NLPGetter; //导入依赖的package包/类
public DemoMultiThread(String modelType, String inputFile, String outputFile, int numThreads) throws Exception
{
AbstractTokenizer tokenizer = NLPGetter.getTokenizer(language);
AbstractComponent tagger = NLPGetter.getComponent(modelType, language, NLPMode.MODE_POS);
AbstractComponent parser = NLPGetter.getComponent(modelType, language, NLPMode.MODE_DEP);
AbstractComponent identifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_PRED);
AbstractComponent classifier = NLPGetter.getComponent(modelType, language, NLPMode.MODE_ROLE);
AbstractComponent labeler = NLPGetter.getComponent(modelType, language, NLPMode.MODE_SRL);
c_components = new AbstractComponent[]{tagger, parser, identifier, classifier, labeler};
process(tokenizer, UTInput.createBufferedFileReader(inputFile), UTOutput.createPrintBufferedFileStream(outputFile), numThreads);
}
开发者ID:clearnlp,项目名称:clearnlp-demo,代码行数:13,代码来源:DemoMultiThread.java
注:本文中的com.clearnlp.nlp.NLPGetter类示例整理自Github/MSDocs等源码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。 |
请发表评论