AI-Game aigame

---

AI-Game aigame is a toolkit for developing and comparing reinforcement learning algorithms.

Basics

There are two basic concepts in reinforcement learning: the environment (namely, the outside world) and the agent (namely, the algorithm you are writing). The agent sends actions to the environment, and the environment replies with observations and rewards (that is, a score).

The core aigame interface is Env <https://github.com/Justontheway/aigame/blob/master/aigame/core.py>_, which is the unified environment interface and agent interface. The following are the Env methods you should know:

• reset(self): Reset the environment's state. Returns observation.
• step(self, action): Step the environment by one timestep. Returns observation, reward, done, info.
• render(self, mode='human', close=False): Render one frame of the environment. The default mode will do something human friendly, such as pop up a window. Passing the close flag signals the renderer to close any such windows.

Installation

On Win7|10

• download ActivePython : http://downloads.activestate.com/ActivePython/releases/2.7.13.2714/ActivePython-2.7.13.2714-win64-x64-402182.exe
• download autopy : https://pypi.python.org/packages/7f/35/a7f1c8c2f2d380c9df73efa56043bf2296e66273074bae37b8625db7b608/autopy-0.51.win-amd64-py2.7.exe#md5=f67dee6f0a30673a2b267dec0431395e
• download aigame : git clone https://github.com/Justontheway/aigame.git
• cd aigame
• python setup.py install
• cd examples
• python random_agent.py

Note

1. 因为运行时需要操作window窗口,所以运行脚本的时候要用管理员身份打开cmd