def __init__(self, context_window_size=4, prime_feature_length=4, add_prime_feature=False, ngram_value=2): # get hold of suitable feature extractor # k_param: pass 4 tokens as context window length: self.feature_extractor = POSContextSequenceFeatureExtractor( k_param=context_window_size, prime_feature_length=prime_feature_length, add_prime_feature=add_prime_feature) self.lexicalized_fe = LexicalizedNgramsFeatureExtractor(ngram_value)
def __init__(self, context_window_size=4, prime_feature_length=4, add_prime_feature = False, ngram_value=2): # get hold of suitable feature extractor # k_param: pass 4 tokens as context window length: self.feature_extractor = POSContextSequenceFeatureExtractor(k_param=context_window_size,prime_feature_length=prime_feature_length, add_prime_feature=add_prime_feature) self.lexicalized_fe = LexicalizedNgramsFeatureExtractor(ngram_value)
class LexicalizedFeatureExtractionFlowMapper: def __init__(self, context_window_size=4, prime_feature_length=4, add_prime_feature=False, ngram_value=2): # get hold of suitable feature extractor # k_param: pass 4 tokens as context window length: self.feature_extractor = POSContextSequenceFeatureExtractor( k_param=context_window_size, prime_feature_length=prime_feature_length, add_prime_feature=add_prime_feature) self.lexicalized_fe = LexicalizedNgramsFeatureExtractor(ngram_value) def process(self, line): try: feature_dict, category, word = self.feature_extractor.extract_features( line) lex_feature_dict, lex_cat, lex_word = self.lexicalized_fe.extract_features( line) # updating the feature dict with lexicalized features as well feature_dict.update(lex_feature_dict) if not feature_dict is None: print word, "\t", category, "\t", feature_dict except Exception as ex: print >> sys.stderr, ex.message pass
class LexicalizedFeatureExtractionFlowMapper: def __init__(self, context_window_size=4, prime_feature_length=4, add_prime_feature = False, ngram_value=2): # get hold of suitable feature extractor # k_param: pass 4 tokens as context window length: self.feature_extractor = POSContextSequenceFeatureExtractor(k_param=context_window_size,prime_feature_length=prime_feature_length, add_prime_feature=add_prime_feature) self.lexicalized_fe = LexicalizedNgramsFeatureExtractor(ngram_value) def process(self, line): try: feature_dict,category,word = self.feature_extractor.extract_features(line) lex_feature_dict, lex_cat, lex_word = self.lexicalized_fe.extract_features(line) # updating the feature dict with lexicalized features as well feature_dict.update(lex_feature_dict) if not feature_dict is None: print word,"\t",category,"\t",feature_dict except Exception as ex: print >>sys.stderr,ex.message pass