Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Hindsight Experience Replay Buffer #84

Open
wants to merge 48 commits into
base: master
Choose a base branch
from

Conversation

prabhatnagarajan
Copy link
Contributor

@prabhatnagarajan prabhatnagarajan commented Oct 28, 2020

Depends on #80.
Resolves #6.

Results:

her_bit_flip_dqn

@prabhatnagarajan
Copy link
Contributor Author

/test

@pfn-ci-bot
Copy link

Successfully created a job for commit 71809a5:

@prabhatnagarajan
Copy link
Contributor Author

/test

@pfn-ci-bot
Copy link

Successfully created a job for commit 9721d1a:

@prabhatnagarajan
Copy link
Contributor Author

/test

@pfn-ci-bot
Copy link

Successfully created a job for commit 8643e0d:

prabhatnagarajan and others added 2 commits October 30, 2020 04:15
Co-authored-by: Justin DuJardin <justindujardin@users.noreply.github.com>
@pfn-ci-bot
Copy link

Successfully created a job for commit 573c7a2:

@prabhatnagarajan
Copy link
Contributor Author

/test

@pfn-ci-bot
Copy link

Successfully created a job for commit ed4ae2e:

@prabhatnagarajan
Copy link
Contributor Author

Current results
PickAndPlace
Push
Slide

@prabhatnagarajan
Copy link
Contributor Author

/test

@pfn-ci-bot
Copy link

Successfully created a job for commit ef230a4:

@ummavi
Copy link
Member

ummavi commented May 20, 2021

/test

@pfn-ci-bot
Copy link

Successfully created a job for commit 35f085c:

success = (
self.observation["desired_goal"] == self.observation["achieved_goal"]
).all()
return (self.steps >= self.n) or success, success
Copy link
Contributor Author

Choose a reason for hiding this comment

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

I think I should use the wrapper ContinuingTimeLimit actually. Exceeding the number of timesteps should not be a terminal signal.

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.

Hindsight Experience Replay
5 participants