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

Add PER #2159

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Add PER #2159

wants to merge 1 commit into from

Conversation

maliesa96
Copy link
Contributor

Closes #1489.

@maliesa96 maliesa96 requested a review from a team as a code owner October 28, 2020 22:39
@maliesa96 maliesa96 requested review from ryanjulian and removed request for a team October 28, 2020 22:39
@mergify mergify bot requested review from a team, yeukfu and ziyiwu9494 and removed request for a team October 28, 2020 22:39
@maliesa96
Copy link
Contributor Author

Benchmarks should be done in a few days.

@codecov
Copy link

codecov bot commented Oct 28, 2020

Codecov Report

Merging #2159 (8f3fb8e) into master (e1c576f) will decrease coverage by 0.01%.
The diff coverage is 88.23%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master    #2159      +/-   ##
==========================================
- Coverage   91.54%   91.52%   -0.02%     
==========================================
  Files         197      198       +1     
  Lines       10927    10979      +52     
  Branches     1387     1393       +6     
==========================================
+ Hits        10003    10049      +46     
- Misses        665      669       +4     
- Partials      259      261       +2     
Impacted Files Coverage Δ
src/garage/envs/gym_env.py 96.21% <ø> (ø)
src/garage/replay_buffer/replay_buffer.py 19.17% <ø> (ø)
src/garage/torch/algos/pearl.py 48.29% <0.00%> (ø)
src/garage/torch/algos/dqn.py 89.85% <61.53%> (-3.23%) ⬇️
src/garage/replay_buffer/per_replay_buffer.py 97.61% <97.61%> (ø)
src/garage/replay_buffer/path_buffer.py 96.07% <100.00%> (+0.07%) ⬆️
src/garage/tf/algos/ddpg.py 97.02% <100.00%> (ø)
src/garage/tf/algos/dqn.py 91.52% <100.00%> (ø)
src/garage/tf/algos/td3.py 96.66% <100.00%> (ø)
src/garage/torch/algos/ddpg.py 96.69% <100.00%> (ø)
... and 3 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e1c576f...8f3fb8e. Read the comment docs.

Copy link
Member

@ryanjulian ryanjulian left a comment

Choose a reason for hiding this comment

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

How are the benchmark results?

Copy link
Contributor

@krzentner krzentner left a comment

Choose a reason for hiding this comment

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

Nice. This is a pretty clean implementation.

src/garage/replay_buffer/per_replay_buffer.py Outdated Show resolved Hide resolved
src/garage/torch/algos/dqn.py Outdated Show resolved Hide resolved
@maliesa96 maliesa96 force-pushed the per_dqn branch 2 times, most recently from 4e708fd to bf2b104 Compare November 6, 2020 19:37
@maliesa96
Copy link
Contributor Author

Currently rerunning some of the benchmarks (some were interrupted), will report back once they're done.

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.

Implement PER
3 participants