I am training an agent for the Cartpole environment of openAI gym, and storing a video, and then I'm trying to display it (doing this in jupyter notebook).
I start with:
env = gym.make('CartPole-v0')
env = wrappers.Monitor(env, "./gym-results", force=True)
Then I do a training loop in the standard way using env.step(action). This causes a video to be stored correctly, but weirdly, it also causes the video to pop-up in realtime (just as in the question Display OpenAI gym in Jupyter notebook only), even though I'm not calling env.render() anywhere. How do I stop this?
