preprocess/circa.py [26:48]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        }

    def get_train_test_lines(self, dataset):
        # only train set, manually split 20% data as test
        map_hf_dataset_to_list = self.get_map_hf_dataset_to_list()
        if map_hf_dataset_to_list is None:
            map_hf_dataset_to_list = self.map_hf_dataset_to_list
        lines = map_hf_dataset_to_list(dataset, "train")

        np.random.seed(42)
        np.random.shuffle(lines)

        n = len(lines)

        train_lines = lines[:int(0.8*n)]
        test_lines = lines[int(0.8*n):]

        return train_lines, test_lines


    def map_hf_dataset_to_list(self, hf_dataset, split_name):
        lines = []
        for datapoint in hf_dataset[split_name]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



preprocess/financial_phrasebank.py [25:46]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        }

    def get_train_test_lines(self, dataset):
        # only train set, manually split 20% data as test
        map_hf_dataset_to_list = self.get_map_hf_dataset_to_list()
        if map_hf_dataset_to_list is None:
            map_hf_dataset_to_list = self.map_hf_dataset_to_list
        lines = map_hf_dataset_to_list(dataset, "train")

        np.random.seed(42)
        np.random.shuffle(lines)

        n = len(lines)

        train_lines = lines[:int(0.8*n)]
        test_lines = lines[int(0.8*n):]

        return train_lines, test_lines

    def map_hf_dataset_to_list(self, hf_dataset, split_name):
        lines = []
        for datapoint in hf_dataset[split_name]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -