基于Unity的强化学习算法DQN的纯C#实现，Sample是一个Cart-Pole环境。
训练刚开始时：
训练200个episode之后：
A simple implementation of the reinforcement learning algorithm DQN in pure C#, and a Cart-Pole environment is used in the sample case.
Beginning of the training:
After training 200+ episodes: