New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

用Atari的pong游戏代替CartPole策略网络会发生错误 #76

Open

wing-on opened this issue Dec 2, 2021 · 1 comment

wing-on commented Dec 2, 2021

我直接把大佬你的PGconfig类中的环境名‘self.env = 'CartPole-v0'，改成 self.env = 'PongDeterministic-v4'，其他没有任何改动，但是出错了。产生了如下错误：RuntimeError: mat1 and mat2 shapes cannot be multiplied (33600x3 and 210x36)。
我看了看觉得是在两个游戏的状态维度不同导致错误，pong的state为（210,160,3），而CartPole-v0的state为（4，），pong在策略网络模型的forward函数向前传递state时就产生错误了。不知道具体错误原因是什么，希望作者大佬能帮忙解答一下。不胜感激！

Contributor

johnjim0816 commented Dec 22, 2021

PongDeterministic-v4

这个环境需要用到卷积神经网络，最近会加

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment