Reinforcement-learning-with-tensorflow
Reinforcement-learning-with-tensorflow copied to clipboard
MorvanZhou
Simple Reinforcement learning tutorials, 莫烦Python 中文AI教学
This is more of a query rather than an issue. So Once the model is trained, how do I validate the model with some random trajectory I provide. So as...
Hi Morvan Zhou, Thanks for the great repo. [Here](https://github.com/tessavdheiden/joint_model/blob/main/train/policy_train_torch.py) is your code, but then in Pytorch. The results between our codes: Tensorflow (your code)  Pytorch (my code) 
Hi! Could you please explain what the replace here does? https://github.com/MorvanZhou/Reinforcement-learning-with-tensorflow/blob/1fd1c08a6c8928d027fb75d51ebf6f9441e3dc33/experiments/Robot_arm/DDPG.py#L101 It somehow passes t_params and e_params to the session..
There are many games in Atari games just like Breakout-v0, SpaceInvaders-ram-v0, and so on. How can I see these games' source code?
教学中的迷宫规模都比较小，不复杂。如果想要求解大规模，如100*100的迷宫，且环境比较复杂的，应该选用什么强化学习算法？我试了几种算法，发现Q-learning貌似求出的不是最优解，而DQN的训练速度太慢，难以求得解。想请问下是什么原因导致的这些问题，随机策略选择还是其他参数设置的问题？或者有什么比较适合的强化学习算法嘛？求大神指导！谢谢！
Please be so kind to delete this issue for me, thank you in advance.
Please be so kind to delete this issue for me, thank you in advance.
Please be so kind to delete this issue for me, thank you in advance.
你好，莫烦老师，我在运行simple_ppo算法中，，根据当前状态选择一个动作 a=self.sess.run(self.sample_op,{self.tfs:s})[0]，，选择出来的动作为nan，，我应该如何修改，才能在运行代码过程中不在出现nan值，
AttributeError: module 'pyglet.clock' has no attribute 'set_fps_limit' @MorvanZhou
Metadata
Owner
Metadata
Simple Reinforcement learning tutorials, 莫烦Python 中文AI教学