def _get_gender_per_word()

in src/morphological_filtering.py [0:0]

• 14 lines of code
• 6 McCabe index (conditional complexity)

    def _get_gender_per_word(self, sentence):
        tokens = set([self.tokenizer(tok).text for tok in sentence.split()])
        gender_per_word = []
        for word in tokens:
            if word != "את":
                if word[-1] in self.fem_chars:
                    gender_per_word.append(FEM_LABEL)
                elif word[-1] in self.msc_chars:
                    gender_per_word.append(MSC_LABEL)
                else:
                    gender_per_word.append(OTHER_LABEL)
            else:
                gender_per_word.append(OTHER_LABEL)
        return gender_per_word