coderserio / chinese-mnist

基于TensorFlow实现的手写中文数字识别项目，包含了完整的Python代码、训练好的模型以及相应的数据集资源。Recognizing handwritten Chinese numerals using TensorFlow, featuring the complete Python code, a pre-trained model, and the associated dataset.

Chinese-MNIST

中文 | English

This project utilizes TensorFlow and Keras to build an image classification model that recognizes Chinese numerals. The repository includes Python source code, a pre-trained model, and the required dataset.

Contents Overview

1. index.py: The main script file containing functionality for data loading, model training, prediction, and evaluation.
2. model/*: Pre-trained model files that can be directly used for prediction and evaluation.
3. dataset/*: Data resources
   • ./chinese_mnist.csv: CSV file containing label information corresponding to the image data.
   • ./data/data/*: Image data directory containing the actual images.