This project is concered with the recogonition competition of a Chinese character dataset, with 100 categories of characters and 400 pictures of each.
The models we used are in model.py and the data processing is completed in data_one.py. TMD_main.py runs the whole training and output process. For more details about the dataset and competition, please refer to https://www.tinymind.cn/competitions/41#dataDescription