private double accuracyFor(InstanceList examples) { TokenAccuracyEvaluator teval = new TokenAccuracyEvaluator(examples, "train"); teval.evaluate(lastTrainer); return teval.getAccuracy("train"); }
public void testTokenAccuracy() { Pipe p = makeSpacePredictionPipe(); InstanceList instances = new InstanceList(p); instances.addThruPipe(new ArrayIterator(data)); InstanceList[] lists = instances.split(new Random(777), new double[] { .5, .5 }); CRF crf = new CRF(p.getDataAlphabet(), p.getTargetAlphabet()); crf.addFullyConnectedStatesForLabels(); CRFTrainerByLabelLikelihood crft = new CRFTrainerByLabelLikelihood(crf); crft.setUseSparseWeights(true); crft.trainIncremental(lists[0]); TokenAccuracyEvaluator eval = new TokenAccuracyEvaluator(lists, new String[] { "Train", "Test" }); eval.evaluateInstanceList(crft, lists[1], "Test"); assertEquals(0.9409, eval.getAccuracy("Test"), 0.001); }
trainer.addEvaluator(new TokenAccuracyEvaluator(testingInstances, "testing")); trainer.train(trainingInstances);
crf.addFullyConnectedStatesForLabels (); CRFTrainerByLabelLikelihood crft = new CRFTrainerByLabelLikelihood (crf); TokenAccuracyEvaluator eval = new TokenAccuracyEvaluator (new InstanceList[] {training, testing}, new String[] {"Training", "Testing"}); for (int i = 0; i < 5; i++) { crft.train (training, 1); eval.evaluate(crft); memm.addFullyConnectedStatesForLabels (); MEMMTrainer memmt = new MEMMTrainer (memm); TransducerEvaluator memmeval = new TokenAccuracyEvaluator (new InstanceList[] {training2, testing2}, new String[] {"Training2", "Testing2"}); memmt.train (training2, 5); memmeval.evaluate(memmt);
public void testTokenAccuracy() { Pipe p = makeSpacePredictionPipe(); InstanceList instances = new InstanceList(p); instances.addThruPipe(new ArrayIterator(data)); InstanceList[] lists = instances.split(new Random(777), new double[] { .5, .5 }); CRF crf = new CRF(p.getDataAlphabet(), p.getTargetAlphabet()); crf.addFullyConnectedStatesForLabels(); CRFTrainerByLabelLikelihood crft = new CRFTrainerByLabelLikelihood(crf); crft.setUseSparseWeights(true); crft.trainIncremental(lists[0]); TokenAccuracyEvaluator eval = new TokenAccuracyEvaluator(lists, new String[] { "Train", "Test" }); eval.evaluateInstanceList(crft, lists[1], "Test"); assertEquals(0.9409, eval.getAccuracy("Test"), 0.001); }
trainer.addEvaluator(new TokenAccuracyEvaluator(testingInstances, "testing")); trainer.train(trainingInstances);
crf.addFullyConnectedStatesForLabels (); CRFTrainerByLabelLikelihood crft = new CRFTrainerByLabelLikelihood (crf); TokenAccuracyEvaluator eval = new TokenAccuracyEvaluator (new InstanceList[] {training, testing}, new String[] {"Training", "Testing"}); for (int i = 0; i < 5; i++) { crft.train (training, 1); eval.evaluate(crft); memm.addFullyConnectedStatesForLabels (); MEMMTrainer memmt = new MEMMTrainer (memm); TransducerEvaluator memmeval = new TokenAccuracyEvaluator (new InstanceList[] {training2, testing2}, new String[] {"Training2", "Testing2"}); memmt.train (training2, 5); memmeval.evaluate(memmt);
public SyllTagModel train(Collection<Alignment> trainInputs, Collection<Alignment> testInputs, boolean eval) { Pipe pipe = makePipe(); InstanceList trainExamples = makeExamplesFromAlignsWithPipe(trainInputs, pipe); InstanceList testExamples = null; if (testInputs != null) { testExamples = makeExamplesFromAlignsWithPipe(testInputs, pipe); } log.info("Training test-time syll aligner on whole data..."); TransducerTrainer trainer = trainOnce(pipe, trainExamples); if (eval) { TokenAccuracyEvaluator evaler = new TokenAccuracyEvaluator(trainExamples, "traindata"); evaler.evaluate(trainer); double trainAcc = evaler.getAccuracy("traindata"); double testAcc = 0.0; if (testExamples != null) { TokenAccuracyEvaluator evaler2 = new TokenAccuracyEvaluator(testExamples, "testdata"); evaler2.evaluate(trainer); testAcc = evaler2.getAccuracy("testdata"); } log.info("Train data accuracy = " + trainAcc + ", test data accuracy = " + testAcc); } return new SyllTagModel((CRF) trainer.getTransducer()); }
trainer.addEvaluator(new TokenAccuracyEvaluator(testingInstances, "testing")); trainer.train(trainingInstances);
eval = new TokenAccuracyEvaluator(new InstanceList[] {trainingData, testData}, new String[] {"Training", "Testing"});
eval = new TokenAccuracyEvaluator(new InstanceList[] {trainingData, testData}, new String[] {"Training", "Testing"}); else if (testOption.value.startsWith("seg="))
eval = new TokenAccuracyEvaluator(new InstanceList[] {trainingData, testData}, new String[] {"Training", "Testing"}); else if (testOption.value.startsWith("seg="))
eval = new TokenAccuracyEvaluator(new InstanceList[] {trainingData, testData}, new String[] {"Training", "Testing"}); else if (testOption.value.startsWith("seg="))
eval = new TokenAccuracyEvaluator(new InstanceList[] {trainingData, testData}, new String[] {"Training", "Testing"});
eval = new TokenAccuracyEvaluator(new InstanceList[] {trainingData, testData}, new String[] {"Training", "Testing"});