def classify(self, featureset):
"""
@return: the most appropriate label for the given featureset.
@rtype: label
"""
if overridden(self.batch_classify):
return self.batch_classify([featureset])[0]
else:
raise NotImplementedError()
def prob_classify(self, featureset):
"""
@return: a probability distribution over labels for the given
featureset.
@rtype: L{ProbDistI <nltk.probability.ProbDistI>}
"""
if overridden(self.batch_prob_classify):
return self.batch_prob_classify([featureset])[0]
else:
raise NotImplementedError()