def __post_init__()

in src/autotrain/preprocessor/text.py [0:0]

• 35 lines of code
• 20 McCabe index (conditional complexity)

    def __post_init__(self):
        # check if text_column, question_column, and answer_column are in train_data
        if self.text_column not in self.train_data.columns:
            raise ValueError(f"{self.text_column} not in train data")
        if self.question_column not in self.train_data.columns:
            raise ValueError(f"{self.question_column} not in train data")
        if self.answer_column not in self.train_data.columns:
            raise ValueError(f"{self.answer_column} not in train data")
        # check if text_column, question_column, and answer_column are in valid_data
        if self.valid_data is not None:
            if self.text_column not in self.valid_data.columns:
                raise ValueError(f"{self.text_column} not in valid data")
            if self.question_column not in self.valid_data.columns:
                raise ValueError(f"{self.question_column} not in valid data")
            if self.answer_column not in self.valid_data.columns:
                raise ValueError(f"{self.answer_column} not in valid data")

        # make sure no reserved columns are in train_data or valid_data
        for column in RESERVED_COLUMNS:
            if column in self.train_data.columns:
                raise ValueError(f"{column} is a reserved column name")
            if self.valid_data is not None:
                if column in self.valid_data.columns:
                    raise ValueError(f"{column} is a reserved column name")

        # convert answer_column to dict
        try:
            self.train_data.loc[:, self.answer_column] = self.train_data[self.answer_column].apply(
                lambda x: ast.literal_eval(x)
            )
        except ValueError:
            logger.warning("Unable to do ast.literal_eval on train_data[answer_column]")
            logger.warning("assuming answer_column is already a dict")

        if self.valid_data is not None:
            try:
                self.valid_data.loc[:, self.answer_column] = self.valid_data[self.answer_column].apply(
                    lambda x: ast.literal_eval(x)
                )
            except ValueError:
                logger.warning("Unable to do ast.literal_eval on valid_data[answer_column]")
                logger.warning("assuming answer_column is already a dict")