CRNN with attention to do OCR,Chinese ocr in Chinese branch
CRNN is base CNN,and BiLSTM with 256 hidden_units is encode network ,GRU with 256 hidden_units is decode network
put your image in 'train' dir,and image name should be like "xx_label_xx.jpg",Parameters are set in config.py,
and then just run the train.py
TensorFlow >=1.2
opencv