def test(self,lstSamples,model): """Test the libSVM classifier on the sample set and return the ClassificationResults.""" classification_results = ClassificationResults() print "Testing on %d samples"%(len(lstSamples)) for sample in lstSamples: #if sample.get_class_label() != self.null_flag: # RVF if True: # RVF. Also classify the sample, if the class is previously unknown best_class = self.classify(sample,model) classification_results.add_classification(sample.id,best_class,sample.get_class_label()) return classification_results
def test(self,lstSamples,model,confusionMatrix={}):
"""Test the CPAR classifier on the sample set and return the ClassificationResults."""
self.rules = model
self.rules.set_target_accuracy(self.parameters["accuracy_measure"])
lstRules = self.rules.remap_feature_to_index(lstSamples)
if len(lstRules) ==0:
return 0
classification_results = ClassificationResults()
print "Testing on %d samples"%(len(lstSamples))
for sample in lstSamples:
#if sample.get_class_label() != self.null_flag: # uncommented by RVF
if True: # RVF. Also classify the sample, if the class is previously unknown
intBestClass = self.classify(sample,lstRules)
classification_results.add_classification(sample.id,intBestClass,sample.get_class_label())
return classification_results
