char_data_iterator.py 文件源码

def __init__(self, source,
                 source_dict,
                 batch_size=128,
                 maxlen=100,
                 minlen=0,
                 n_words_source=-1):
        if source.endswith('.gz'):
            self.source = gzip.open(source, 'r')
        else:
            self.source = open(source, 'r')

        self.source_dict = {'1': 1, '0': 0, 0: 3}
        self.batch_size = batch_size
        self.maxlen = maxlen
        self.minlen = 10
        self.n_words_source = n_words_source

        self.end_of_data = False