def _prepare_identity_terms()

in data_measurements/npmi/npmi.py [0:0]

• 13 lines of code
• 7 McCabe index (conditional complexity)

    def _prepare_identity_terms(self):
        """Uses DataFrame magic to return those terms that appear
        greater than min_vocab times."""
        # Mask to get the identity terms
        true_false = [term in self.vocab_counts_df.index for term in
                      self.identity_terms]
        # List of identity terms
        word_list_tmp = [x for x, y in zip(self.identity_terms, true_false) if
                         y]
        # Whether said identity terms have a count > min_count
        true_false_counts = [
            self.vocab_counts_df.loc[word, CNT] >= self.min_count for word in
            word_list_tmp]
        # List of identity terms with a count higher than min_count
        avail_identity_terms = [word for word, y in
                                zip(word_list_tmp, true_false_counts) if y]
        logs.debug("Identity terms that occur > %s times are:" % self.min_count)
        logs.debug(avail_identity_terms)
        return avail_identity_terms