Gym: What's the meaning of the input of CartPole observation space?

Created on 7 Jul 2016 · 4Comments · Source: openai/gym

Hello, all, i'm newbie to gym.
I print out the env.observation_space.shape[0], and it equals 4(CartPole-v0 env), so What's the meaning of this 4 numbers,? i cannot found the doc which describe it.( i think it may include the position of cart, the angle of the pole, the speed of the cart and the speed of the pole.)
Thanks!

Source

tornadomeet

👍15

Most helpful comment

Good general-purpose agents don't need to know the semantics of the observations: they can learn how to map observations to actions to maximize reward without any prior knowledge.

That said, it's:
[position of cart, velocity of cart, angle of pole, rotation rate of pole]. Defined at https://github.com/openai/gym/blob/master/gym/envs/classic_control/cartpole.py#L75

tlbtlbtlb on 7 Jul 2016

👍50 🎉1

All 4 comments

Good general-purpose agents don't need to know the semantics of the observations: they can learn how to map observations to actions to maximize reward without any prior knowledge.

That said, it's:
[position of cart, velocity of cart, angle of pole, rotation rate of pole]. Defined at https://github.com/openai/gym/blob/master/gym/envs/classic_control/cartpole.py#L75

tlbtlbtlb on 7 Jul 2016

👍50 🎉1

great!

tornadomeet on 8 Jul 2016

Is there a documentation which explains what the gym observation is returning. Please Share a Link.