INNER CODE UNIT · Python

load_data

shibing624/pytextclassifier · pytextclassifier/base_classifier.py:12

def load_data(data_list_or_path, header=None, names=('labels', 'text'), delimiter='\t',
              labels_sep=',', is_train=False):
    """
    Encoding data_list text
    @param data_list_or_path: list of (label, text), eg: [(label, text), (label, text) ...]
    @param header: read_csv header
    @param names: read_csv names
    @param delimiter: read_csv sep
    @param labels_sep: multi label split
    @param is_train: is train data
    @return: X, y, data_df
    """
    if isinstance(data_list_or_path, list):
        data_df = pd.DataFrame(data_list_or_path, columns=names)
    elif isinstance(data_list_or_path, str) and os.path.exists(data_list_or_path):
        data_df = pd.read_csv(data_list_or_path, header=header, delimiter=delimiter, names=names)
    elif isinstance(data_list_or_path, pd.DataFrame):
        data_df = data_list_or_path

View source record →