Skip to content

Conversation

younik
Copy link
Member

@younik younik commented Apr 7, 2023

Adding a draft for ReplayBuffer

assert observation_space.shape is not None
assert action_space.shape is not None

self._obs = np.zeros((self.max_size, *observation_space.shape), dtype=observation_space.dtype)
Copy link
Contributor

@Howuhh Howuhh Apr 7, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think in more general implementation, buffer should use here gym.spaces.unflatten to provide observations samples in their original shape

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@Howuhh Thanks for the review; I am working on it!

Are you in our Discord channel? Can you drop me a message? (younik)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we make the unflatten optional?

@younik younik closed this Jun 1, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants