Java PrintInputAndTarget类代码示例

import cc.mallet.pipe.PrintInputAndTarget; //导入依赖的package包/类
public void testPrint() {
	Pipe p = new SerialPipes(new Pipe[] {
			new CharSequence2TokenSequence("."), new TokenText(),
			new TestCRFTokenSequenceRemoveSpaces(),
			new TokenSequence2FeatureVectorSequence(),
			new PrintInputAndTarget(), });
	InstanceList one = new InstanceList(p);
	String[] data = new String[] { "ABCDE", };
	one.addThruPipe(new ArrayIterator(data));
	CRF crf = new CRF(p, null);
	crf.addFullyConnectedStatesForThreeQuarterLabels(one);
	CRFTrainerByLabelLikelihood crft = new CRFTrainerByLabelLikelihood(crf);
	crf.setWeightsDimensionAsIn(one, false);
	Optimizable mcrf = crft.getOptimizableCRF(one);
	double[] params = new double[mcrf.getNumParameters()];
	for (int i = 0; i < params.length; i++) {
		params[i] = i;
	}
	mcrf.setParameters(params);
	crf.print();
}
 

import cc.mallet.pipe.PrintInputAndTarget; //导入依赖的package包/类
public void disabledtestPrint ()
{
	Pipe p = new SerialPipes (new Pipe[] {
     new CharSequence2TokenSequence("."),
		 new TokenText(),
		 new TestMEMM.TestMEMMTokenSequenceRemoveSpaces(),
		 new TokenSequence2FeatureVectorSequence(),
		 new PrintInputAndTarget(),
  });
	InstanceList one = new InstanceList (p);
	String[] data = new String[] { "ABCDE", };
	one.addThruPipe (new ArrayIterator (data));
	MEMM crf = new MEMM (p, null);
	crf.addFullyConnectedStatesForLabels();
	crf.setWeightsDimensionAsIn (one);
	MEMMTrainer memmt = new MEMMTrainer (crf);
	MEMMTrainer.MEMMOptimizableByLabelLikelihood mcrf = memmt.getOptimizableMEMM(one);
	double[] params = new double[mcrf.getNumParameters()];
	for (int i = 0; i < params.length; i++) {
		params [i] = i;
	}
	mcrf.setParameters (params);
	crf.print ();
}
 

import cc.mallet.pipe.PrintInputAndTarget; //导入依赖的package包/类
public static void testMultiTagSerialization () throws IOException, ClassNotFoundException
{
  Pipe origPipe = new SerialPipes (new Pipe[] {
          new SimpleTaggerSentence2TokenSequence (),
          new TokenText (),
          new RegexMatches ("digits", Pattern.compile ("[0-9]+")),
          new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")),
          new OffsetFeatureConjunction ("time",
                  new String[] { "digits", "ampm" },
                  new int[] { 0, 1 },
                  true),
          new PrintInputAndTarget (),
  });

  Pipe mtPipe = (Pipe) TestSerializable.cloneViaSerialization (origPipe);
  InstanceList mtLst = new InstanceList (mtPipe);
  mtLst.addThruPipe (new ArrayIterator (doc1));
  Instance mtInst = mtLst.get (0);
  TokenSequence mtTs = (TokenSequence) mtInst.getData ();
  assertEquals (6, mtTs.size ());
  assertEquals (1.0, mtTs.get (3).getFeatureValue ("time"), 1e-15);
  assertEquals (1.0, mtTs.get (4).getFeatureValue ("time"), 1e-15);
}
 

import cc.mallet.pipe.PrintInputAndTarget; //导入依赖的package包/类
public void disabledtestPrint ()
{
	Pipe p = new SerialPipes (new Pipe[] {
     new CharSequence2TokenSequence("."),
		 new TokenText(),
		 new TestMEMMTokenSequenceRemoveSpaces(),
		 new TokenSequence2FeatureVectorSequence(),
		 new PrintInputAndTarget(),
  });
	InstanceList one = new InstanceList (p);
	String[] data = new String[] { "ABCDE", };
	one.addThruPipe (new ArrayIterator (data));
	MEMM crf = new MEMM (p, null);
	crf.addFullyConnectedStatesForLabels();
	crf.setWeightsDimensionAsIn (one);
	MEMMTrainer memmt = new MEMMTrainer (crf);
	MEMMTrainer.MEMMOptimizableByLabelLikelihood mcrf = memmt.getOptimizableMEMM(one);
	double[] params = new double[mcrf.getNumParameters()];
	for (int i = 0; i < params.length; i++) {
		params [i] = i;
	}
	mcrf.setParameters (params);
	crf.print ();
}
 

import cc.mallet.pipe.PrintInputAndTarget; //导入依赖的package包/类
public void testSpacePipe () {
	Pipe p = new SerialPipes(new Pipe[] {
		new CharSequence2TokenSequence("."),
		new TokenSequenceLowercase(),
		new TestCRF.TestCRFTokenSequenceRemoveSpaces (),
		new TokenText(),
		new OffsetConjunctions(false,
														new int[][] {{0},
																				 {1},{-1,0},{0,1},
																				 {-2,-1,0}, {0,1,2}, {-3,-2,-1}, {1,2,3},
														}),
		new PrintInputAndTarget(),
	});

	// Print to a string
	ByteArrayOutputStream out = new ByteArrayOutputStream ();
	PrintStream oldOut = System.out;
	System.setOut (new PrintStream (out));

	InstanceList lst = new InstanceList (p);
	lst.addThruPipe (new ArrayIterator(new String[] { TestCRF.data[0],
																						 TestCRF.data[1], }));

	System.setOut (oldOut);
	
	assertEquals (spacePipeOutput, out.toString());
}
 

import cc.mallet.pipe.PrintInputAndTarget; //导入依赖的package包/类
public void testCopyStatesAndWeights() {
	Pipe p = new SerialPipes(new Pipe[] {
			new CharSequence2TokenSequence("."), new TokenText(),
			new TestCRFTokenSequenceRemoveSpaces(),
			new TokenSequence2FeatureVectorSequence(),
			new PrintInputAndTarget(), });
	InstanceList one = new InstanceList(p);
	String[] data = new String[] { "ABCDE", };
	one.addThruPipe(new ArrayIterator(data));
	CRF crf = new CRF(p, null);
	crf.addFullyConnectedStatesForLabels();
	CRFTrainerByLabelLikelihood crft = new CRFTrainerByLabelLikelihood(crf);
	crf.setWeightsDimensionAsIn(one, false);
	Optimizable.ByGradientValue mcrf = crft.getOptimizableCRF(one);
	double[] params = new double[mcrf.getNumParameters()];
	for (int i = 0; i < params.length; i++) {
		params[i] = i;
	}
	mcrf.setParameters(params);

	StringWriter out = new StringWriter();
	crf.print(new PrintWriter(out, true));
	System.out.println("------------- CRF1 -------------");
	crf.print();

	// Make a copy of this CRF
	CRF crf2 = new CRF(crf);

	StringWriter out2 = new StringWriter();
	crf2.print(new PrintWriter(out2, true));
	System.out.println("------------- CRF2 -------------");
	crf2.print();

	assertEquals(out.toString(), out2.toString());

	double val1 = mcrf.getValue();
	CRFTrainerByLabelLikelihood crft2 = new CRFTrainerByLabelLikelihood(
			crf2);
	double val2 = crft2.getOptimizableCRF(one).getValue();
	assertEquals(val1, val2, 1e-5);
}
 

import cc.mallet.pipe.PrintInputAndTarget; //导入依赖的package包/类
public static void testMultiTag ()
{
  Pipe mtPipe = new SerialPipes (new Pipe[] {
          new SimpleTaggerSentence2TokenSequence (),
          new TokenText (),
          new RegexMatches ("digits", Pattern.compile ("[0-9]+")),
          new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")),
          new OffsetFeatureConjunction ("time",
                  new String[] { "digits", "ampm" },
                  new int[] { 0, 1 },
                  true),
          new PrintInputAndTarget (),
  });
  Pipe noMtPipe = new SerialPipes (new Pipe[] {
          new SimpleTaggerSentence2TokenSequence (),
          new TokenText (),
          new RegexMatches ("digits", Pattern.compile ("[0-9]+")),
          new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")),
          new OffsetFeatureConjunction ("time",
                  new String[] { "digits", "ampm" },
                  new int[] { 0, 1 },
                  false),
          new PrintInputAndTarget (),
  });

  InstanceList mtLst = new InstanceList (mtPipe);
  InstanceList noMtLst = new InstanceList (noMtPipe);

  mtLst.addThruPipe (new ArrayIterator (doc1));
  noMtLst.addThruPipe (new ArrayIterator (doc1));

  Instance mtInst = mtLst.get (0);
  Instance noMtInst = noMtLst.get (0);

  TokenSequence mtTs = (TokenSequence) mtInst.getData ();
  TokenSequence noMtTs = (TokenSequence) noMtInst.getData ();

  assertEquals (6, mtTs.size ());
  assertEquals (6, noMtTs.size ());

  assertEquals (1.0, mtTs.get (3).getFeatureValue ("time"), 1e-15);
  assertEquals (1.0, noMtTs.get (3).getFeatureValue ("time"), 1e-15);
  assertEquals (1.0, mtTs.get (4).getFeatureValue ("time"), 1e-15);
  assertEquals (0.0, noMtTs.get (4).getFeatureValue ("time"), 1e-15);
}