Skip to content

Reinforcement Tutorial (DQN) #1432

Closed
@sambaPython24

Description

@sambaPython24

Hey,
I try to reproduce PyTorch Reinforcement Tutorial (DGN) in

In each time step, the state of the environment need to be evaluated in the function def get_screen()

The line

screen = env.render(mode='rgb_array').transpose((2, 0, 1))

throws an error both in Google Colabs and on my local machine. The error is related to the gym environment

env = gym.make('CartPole-v0').unwrapped

Is there any idea, how to solve this problem and make this tutorial reproducible again?

Metadata

Metadata

Assignees

No one assigned

    Labels

    Reinforcement LearningIssues relating to reinforcement learning tutorials

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions