google-compute-engine - 使用谷歌计算引擎运行 openai gym

Question

我试图在 Google Compute Engine VM 上运行 CartPole-v0 示例。 https://gym.openai.com/docs

import gym
env = gym.make('CartPole-v0')
env.reset()
for _ in range(1000):
    env.render()
    env.step(env.action_space.sample()) # take a random action

我安装了 XQuartz，并使用 -X 进行 ssh。

首先，我得到pyglet.canvas.xlib.NoSuchDisplayException:无法连接到“无”

然后根据http://www.gitterforum.com/discussion/openai-gym?page=28 , 我使用 "xvfb-run -s "-screen 0 1400x900x24"/bin/bash"

然后它运行良好，结果:

Making new env: CartPole-v0
You are calling 'step()' even though this environment has already returned done = True. You should always call 'reset()' once you receive 'done = True' -- any further steps are undefined behavior

但是什么都没有渲染...

Answer 1

使用 https://gym.openai.com/docs 上观察部分的代码.