rl_channel_allocation-master
所属分类:其他
开发工具:Python
文件大小:5KB
下载次数:10
上传日期:2019-03-20 21:26:56
上 传 者:
enchanting
说明: 有关于D2D通信中,信道仿真的编程是基于python语言的。
(Programming for channel emulation in D2D communication)
文件列表:
environment.py (6143, 2018-09-13)
qlearn.py (2038, 2018-09-13)
requirements.txt (3227, 2018-09-13)
# rl_channel_allocation
Basic info
This is essentially a bin packing problem environment. It is in it's early stages of development but is fully operational.
There are two python scripts in this repistory:
1) environment.py : This is the env which the agent exists in and implements selected actions and calculates the reward earned.
2) qlearn.py : This is the Q-Learning algorithm and handles the acquisition of the calculation of the state-action values.
TODO:
- Enable larger state spaces.
- Allow for dynamic figuring out of actions.
- Find what the best learning rate and discount factor are for this problem.
- Add prioritisation in.
- Add time-awareness.
近期下载者:
相关文件:
收藏者: