# Branching DQN
This repo holds an implementation of a PyTorch version of Branching Double Deep Q-Learning. It is based on my other repo for Double DQN: https://github.com/mome36/DoubleDQN and on the paper https://arxiv.org/pdf/1711.08946.pdf
Branching DQN allows a Q-Learning agent to select multiple actions simultaneously. It can be used for continuous control, as I've demonstrated in `BipedalWalker-v2` or for controlling a video game character (yielding actions sur as Jump and Hit instead of Jump or Hit).
## How to use:
To train an agent, run:
```bash
python dueling_dqn.py --env BipedalWalker-v2
```
To see the agent perform:
```bash
python enjoy.py --env BipedalWalker-v2
```
## Performances
基于python深度强化学习控制的节能热模拟加热系统设计.zip （51个子文件） 
code1128 项目图片 
项目图片 (1).pdf 198KB
项目图片 (1).png 77KB 项目图片 (4).png 181KB 项目图片 (3).png 198KB
项目图片 (2).jpg 67KB 项目图片 (4).jpg 71KB 项目图片 (2).png 856KB 项目图片 (1).jpg 64KB 项目图片 (3).jpg 208KB
串口数据读取.py 504B 专利1：一种高速飞行器快速均匀静载加热装置和控制方法.pdf 278KB
热模拟 (2).avi 1.25MB
reward01.txt 1KB 软件著作2：基于蒙特卡洛光追踪法的计算火箭发动机壳体多层石英灯阵列辐射加热热流密度计算软件.jpg 1.71MB 热模拟 (1).avi 2.05MB 热模拟 (3).avi 1.68MB model project_V2.py 24KB project_V5.py 25KB project_V4.py 18KB 还没改完的奖励函数.py 24KB project.py 22KB 文件读取.py 378B project_v5.ckpt 2.49MB project_V3.py 25KB project2_V1.py 21KB test.py 10KB project_V6_exp.py 18KB DATA.txt 6.18MB get_reward 【实时图像+只考虑温差的reward+分段线性】.txt 2KB 3Dpython程序 surface pure data.txt 942KB transist2.txt 61B transist1.txt 4KB C++程序.txt 22KB surfacee tempureture distribution.txt 1.15MB 3D.py 17KB DATA.txt 6.18MB BranchingDQN-master utils.py 3KB runs BipedalWalker-v2 rewards.csv 13KB model_state_dict 92KB reward.png 42KB BipedalWalker-v3 rewards.csv 279B model_state_dict 92KB reward.png 34KB model.py 2KB branching_dqn.py 3KB enjoy.py 1KB __pycache__ model.cpython-36.pyc 2KB utils.cpython-36.pyc 5KB README.md 774B README.md 105B 软件著作1：基于自动化深度强化学习的火箭发动机模型加热控制软件(1).jpg 1.7MB资源评论
