-1
$\begingroup$

I haven't been able to understand the output that OpenAI gym return for observation from this snippet

env = gym.make('ALE/Breakout-ram-v5', render_mode='human', obs_type='grayscale')
obv = env.reset()
print(obv)

enter image description here

from the OpenAI documentation it suggest it return the pixel data from camera, I am guessing that it a the game state representation at a given step. (I might be wrong here, I am still new to OpenAI)

Assuming if it is the pixel data from camera, How can I retrieve information like where the paddle position, the ball velocity (incl. direction) such that it satisfy Markov Properties for Q-Learning

$\endgroup$
1
  • $\begingroup$ Your question is off-topic here. Asking about software is off-topic here. See our on-topic page: ai.stackexchange.com/help/on-topic. $\endgroup$
    – nbro
    Mar 19, 2022 at 21:06

1 Answer 1

0
$\begingroup$

Founded Documentation on Mapping Annotation: Mapping of RAM indexes to semantic state variables from mila-iqia

https://github.com/mila-iqia/atari-representation-learning/blob/master/atariari/benchmark/ram_annotations.py

$\endgroup$

Not the answer you're looking for? Browse other questions tagged .