如何以与FSNS数据集相同的格式创建数据集，以便进行注意力\u ocr

Question

我在How to create dataset in the same format as the FSNS dataset？中尝试了答案代码

对于图像类，我得到以下错误：

File "createFsns.py", line 101, in 'image/class': _int64_feature(char_ids_padded), File "createFsns.py", 
line 52, in _int64_feature return tf.train.Feature(int64_list=tf.train.Int64List(value=[value]))
TypeError: [20, 10, 0, 2, 10, 15, 0, 1, 1, 3] has type list, but expected one of: int, long

我将tf.train.Feature(int64_list=tf.train.Int64List(value=[value])更正为tf.train.Feature(int64_list=tf.train.Int64List(value=value)，现在得到以下错误：

tf.train.Feature(int64_list=tf.train.Int64List(value=value)) TypeError: Value must be iterable