def __init__()

in parlai/tasks/wizard_of_wikipedia/agents.py [0:0]

• 113 lines of code
• 8 McCabe index (conditional complexity)

    def __init__(self, opt, shared=None):
        super().__init__(opt, shared)

        # get number of examples
        self.num_exs = 0
        for ep in range(self.num_episodes()):
            d = self.data[ep]
            for entry in d['dialog']:
                if (
                    entry.get('checked_sentence', None) is not None
                    and entry.get('checked_sentence') != {}
                    and TOKEN_NOCHOSEN not in entry.get('checked_sentence')
                ):
                    self.num_exs += 1
        self.stop_words = [
            'i',
            'a',
            'an',
            'am',
            'are',
            'about',
            'as',
            'at',
            'be',
            'by',
            'for',
            'from',
            'how',
            'in',
            'is',
            'it',
            'of',
            'on',
            'or',
            'that',
            'the',
            'this',
            'to',
            'was',
            'what',
            'when',
            'where',
            '--',
            '?',
            '.',
            "''",
            "''",
            "``",
            ',',
            'do',
            'see',
            'want',
            'people',
            'and',
            "n't",
            "me",
            'too',
            'own',
            'their',
            '*',
            "'s",
            'not',
            'than',
            'other',
            'you',
            'your',
            'know',
            'just',
            'but',
            'does',
            'really',
            'have',
            'into',
            'more',
            'also',
            'has',
            'any',
            'why',
            'will',
            'with',
            'well',
            'still',
            'he',
            'she',
            'we',
            'may',
            'these',
            'his',
            'hers',
            'which',
            'such',
            'they',
            'its',
            'were',
            'my',
            'there',
            ';',
            '-',
            ':',
            '|',
            '&',
            ')',
            '(',
        ]

        try:
            import nltk
        except ImportError:
            raise ImportError('Please install nltk (e.g. pip install nltk).')
        # nltk-specific setup
        st_path = 'tokenizers/punkt/{0}.pickle'.format('english')
        try:
            self.sent_tok = nltk.data.load(st_path)
        except LookupError:
            nltk.download('punkt')
            self.sent_tok = nltk.data.load(st_path)

        self.teacher_type = opt.get('teacher_type')