def _check_data(self):
        if self.dataset_type not in ("TRAIN", "TEST"):
            raise Exception("Dataset type is not valid: 'TRAIN' or 'TEST'")

        if self.dataset_type == "TRAIN":
            if len(self.labels) == 0:
                raise Exception("Label are not found for TRAIN SET")
            if len(self.texts) != len(self.labels):
                raise Exception("Texts and labels are different size.")

        elif self.dataset_type == "TEST":
            self.labels = None

        print(f"Generate {self.dataset_type} SET")
        print(f"output text shape: ({self.data_size}, {self.max_len})")

    def _encode(self, text_chunk):
        encoding_texts = self.tokenizer(
            text_chunk.tolist(),
            return_token_type_ids=False,
            return_attention_mask=True,
            padding="max_length",
            max_length=self.max_len,
            truncation=True,
        )