Pass env_id to replay buffer methods to correctly support batch training #442
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
to correctly handles when episodes end in batch training
muupan
changed the title
muupan
changed the title
|
I checked the effect of the bug fixed by this PR. I added the Before this PR,
|
muupan
changed the title
prabhatnagarajan
suggested changes
May 22, 2019
| # It should have: | ||
| # - 4 transitions from env_id=1 | ||
| # - 5 transitions from env_id=2 | ||
| self.assertEqual(len(rbuf), 9) |
There was a problem hiding this comment.
Shouldn't we have 3 transitions? One from env_id=1 and five from env_id=2?
There was a problem hiding this comment.
Four from env_id=1: (s_0, s_4), (s_1, s_4), (s_2, s_4), and (s_3, s_4), since the transition to s_4 is terminal.
prabhatnagarajan
approved these changes
May 23, 2019
| # It should have: | ||
| # - 4 transitions from env_id=1 | ||
| # - 5 transitions from env_id=2 | ||
| self.assertEqual(len(rbuf), 9) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Merge #443 before this PR.Current replay buffers with num_steps > 1 or episodic are not correct in batch training because they cannot know which env a given transition came from.
This PR adds the
env_idargument to two methods of replay buffers:appendandstop_current_episode. Fromenv_idreplay buffers can know which env a given transition came from and which env's episode is stopped.TODO: