def transform_one(self, obs, target, id):
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     target_tokens = nlp_utils._tokenize(target, token_pattern)
     obs_ngrams = ngram_utils._ngrams(obs_tokens, self.ngram)
     target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
     pos_list = _inter_norm_pos_list(obs_ngrams, target_ngrams)
     return pos_list

 def transform_one(self, obs, target, id):
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     target_tokens = nlp_utils._tokenize(target, token_pattern)
     obs_ngrams = ngram_utils._ngrams(obs_tokens, self.ngram)
     target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
     s = 0.
     for w1 in obs_ngrams:
         for w2 in target_ngrams:
             if dist_utils._is_str_match(w1, w2, self.str_match_threshold):
                 s += 1.
     return np_utils._try_divide(s, len(obs_ngrams)*len(target_ngrams))

 def transform_one(self, obs, target, id):
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     target_tokens = nlp_utils._tokenize(target, token_pattern)
     obs_ngrams = ngram_utils._ngrams(obs_tokens, self.ngram)
     target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
     val_list = []
     for w1 in obs_ngrams:
         s = 0.
         for w2 in target_ngrams:
             if dist_utils._is_str_match(w1, w2, self.str_match_threshold):
                 s += 1.
         val_list.append(np_utils._try_divide(s, len(target_ngrams)))
     if len(val_list) == 0:
         val_list = [config.MISSING_VALUE_NUMERIC]
     return val_list

 def transform_one(self, obs, target, id):
     df = self.dfTrain[self.dfTrain["search_term"] == obs].copy()
     val_list = [config.MISSING_VALUE_NUMERIC]
     if df is not None:
         df = df[df["id"] != id].copy()
         df = df[df["relevance"] == self.relevance].copy()
         if df is not None and df.shape[0] > 0:
             target_tokens = nlp_utils._tokenize(target, token_pattern)
             target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
             val_list = []
             for x in df[self.target_field]:
                 x_tokens = nlp_utils._tokenize(x, token_pattern)
                 x_ngrams = ngram_utils._ngrams(x_tokens, self.ngram)
                 val_list.append(dist_utils._jaccard_coef(x_ngrams, target_ngrams))
     return val_list

 def transform_one(self, obs, target, id):
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     counter = Counter(obs_tokens)
     count = np.asarray(list(counter.values()))
     proba = count/np.sum(count)
     entropy = -np.sum(proba*np.log(proba))
     return entropy

 def _get_avg_ngram_doc_len(self):
     lst = []
     for target in self.target_corpus:
         target_tokens = nlp_utils._tokenize(target, token_pattern)
         target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
         lst.append(len(target_ngrams))
     return np.mean(lst)

 def __iter__(self):
     for column in self.columns:
         for sentence in self.df[column]:
             if not sentence in self.sent_label:
                 self.cnt += 1
                 self.sent_label[sentence] = "SENT_%d"%self.cnt
             tokens = nlp_utils._tokenize(sentence, token_pattern)
             yield LabeledSentence(words=tokens, tags=[self.sent_label[sentence]])

 def transform_one(self, obs, target, id):
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     target_tokens = nlp_utils._tokenize(target, token_pattern)
     obs_ngrams = ngram_utils._ngrams(obs_tokens, self.ngram)
     target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
     val_list = []
     for w1 in obs_ngrams:
         _val_list = []
         for w2 in target_ngrams:
             s = dist_utils._edit_dist(w1, w2)
             _val_list.append(s)
         if len(_val_list) == 0:
             _val_list = [ config.MISSING_VALUE_NUMERIC ]
         val_list.append( _val_list )
     if len(val_list) == 0:
         val_list = [ [config.MISSING_VALUE_NUMERIC] ]
     return val_list

 def transform_one(self, obs, target, id):
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     target_tokens = nlp_utils._tokenize(target, token_pattern)
     obs_synset_list = [wn.synsets(obs_token) for obs_token in obs_tokens]
     target_synset_list = [wn.synsets(target_token) for target_token in target_tokens]
     val_list = []
     for obs_synset in obs_synset_list:
         _val_list = []
         for target_synset in target_synset_list:
             _s = self._maximum_similarity_for_two_synset_list(obs_synset, target_synset)
             _val_list.append(_s)
         if len(_val_list) == 0:
             _val_list = [config.MISSING_VALUE_NUMERIC]
         val_list.append( _val_list )
     if len(val_list) == 0:
         val_list = [[config.MISSING_VALUE_NUMERIC]]
     return val_list

 def transform_one(self, obs, target, id):
     val_list = []
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     target_tokens = nlp_utils._tokenize(target, token_pattern)
     for obs_token in obs_tokens:
         _val_list = []
         if obs_token in self.model:
             for target_token in target_tokens:
                 if target_token in self.model:
                     sim = dist_utils._cosine_sim(self.model[obs_token], self.model[target_token]) 
                     _val_list.append(sim)
         if len(_val_list) == 0:
             _val_list = [config.MISSING_VALUE_NUMERIC]
         val_list.append( _val_list )
     if len(val_list) == 0:
         val_list = [[config.MISSING_VALUE_NUMERIC]]
     return val_list

 def _get_df_dict(self):
     # smoothing
     d = defaultdict(lambda : 1)
     for target in self.target_corpus:
         target_tokens = nlp_utils._tokenize(target, token_pattern)
         target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
         for w in set(target_ngrams):
             d[w] += 1
     return d

 def transform_one(self, obs, target, id):
     obs_tokens = nlp_utils._tokenize(obs, token_pattern)
     target_tokens = nlp_utils._tokenize(target, token_pattern)
     obs_ngrams = ngram_utils._ngrams(obs_tokens, self.ngram)
     target_ngrams = ngram_utils._ngrams(target_tokens, self.ngram)
     return np_utils._try_divide(self._get_match_count(obs_ngrams, target_ngrams, self.idx), len(target_ngrams))