# muzero_dc

**Repository Path**: SpongeBob1008/muzero_dc

## Basic Information

- **Project Name**: muzero_dc
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 1
- **Forks**: 0
- **Created**: 2022-03-26
- **Last Updated**: 2022-05-18

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# muzero_dc

#### 介绍

- 该项目为大创项目，基于MuZero框架的强化学习系统，可适配不同游戏、神经网络。
- 启动在`Launcher.py`，训练、调试可通过`MuZero`类。
- DQN_CNN是用卷积神经网络训练的强化学习系统，代码中只用作tank游戏的训练
- 因比赛仓促，代码比较乱，如有问题可**邮件**咨询，希望对你有帮助！
- [给我发邮件](mailto:19726006@bjtu.edu.cn)(19726006@bjtu.edu.cn)


#### 项目组成员

1.  李明翀
2.  田芸嘉
3.  肖云轩

#### Introduction

- This project is a competition project. A reinforcement learning system based on the MuZero framework can be adapted to different games and neural networks.
- Luanch by `Launcher.py`, and train or debug by the class `MuZero`.
- DQN_CNN is a reinforcement learning system trained with a convolutional neural network, and the code is only used for the training of tank games
- Due to the rush of the competition, the code is messy, if you have any questions, you can **mail me** for consultation, I hope it will help you!
- [Mail to me](mailto:19726006@bjtu.edu.cn)(19726006@bjtu.edu.cn)


#### Members

1.  Li Mingchong
2.  Tian Yunjia
3.  Xiao Yunxuan