eliascai/idcard-ocr

端到端的针对身份证的文字识别

idcard-ocr's Issues

您好，请问一下，用你提供的在corpus下的语料直接训练会报错，请问是语料行数的问题吗？

........................................

2019-08-27 20:22:56 Epoch 1/1000
2019-08-27 20:22:56 Exception in thread Thread-2:
2019-08-27 20:22:56 Traceback (most recent call last):
2019-08-27 20:22:56 File "/usr/local/anaconda3/lib/python3.6/threading.py", line 916, in _bootstrap_inner
2019-08-27 20:22:56 self.run()
2019-08-27 20:22:56 File "/usr/local/anaconda3/lib/python3.6/threading.py", line 864, in run
2019-08-27 20:22:56 self._target(*self._args, **self._kwargs)
2019-08-27 20:22:56 File "/usr/local/anaconda3/lib/python3.6/site-packages/keras/utils/data_utils.py", line 630, in data_generator_task
2019-08-27 20:22:56 generator_output = next(self._generator)
2019-08-27 20:22:56 ValueError: generator already executing
2019-08-27 20:22:56
........................

2019-08-27 20:22:56 The number of corpus is 84
2019-08-27 20:22:56 Exception in thread Thread-1:
2019-08-27 20:22:56 ValueError: Sample larger than population or is negative