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

Use max episode length eval in SAC evaluation #1908

Merged
merged 1 commit into from Aug 20, 2020

Conversation

avnishn
Copy link
Member

@avnishn avnishn commented Aug 16, 2020

closes #1853

@avnishn avnishn requested a review from a team August 16, 2020 01:23
@avnishn avnishn requested a review from a team as a code owner August 16, 2020 01:23
@avnishn avnishn requested review from ziyiwu9494, maliesa96 and krzentner and removed request for a team August 16, 2020 01:23
@mergify mergify bot requested a review from a team August 16, 2020 01:24
@avnishn avnishn linked an issue Aug 16, 2020 that may be closed by this pull request
@ryanjulian ryanjulian added ready-to-merge backport-to-2020.06 Backport this PR to release-2020.06 labels Aug 17, 2020
@ryanjulian ryanjulian changed the title Use max episode length eval in sac evaluation Use max episode length eval in SAC evaluation Aug 19, 2020
@ryanjulian
Copy link
Member

@Mergifyio rebase

@mergify
Copy link
Contributor

mergify bot commented Aug 19, 2020

Command rebase: success

Branch has been successfully rebased

@ahtsan ahtsan force-pushed the avnish-use-max-episode-length-eval branch from 2743e1e to ccf66fc Compare August 19, 2020 18:17
@codecov
Copy link

codecov bot commented Aug 19, 2020

Codecov Report

Merging #1908 into master will decrease coverage by 0.02%.
The diff coverage is 100.00%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master    #1908      +/-   ##
==========================================
- Coverage   93.45%   93.42%   -0.03%     
==========================================
  Files         192      192              
  Lines       10180    10180              
  Branches     1267     1267              
==========================================
- Hits         9514     9511       -3     
- Misses        439      441       +2     
- Partials      227      228       +1     
Impacted Files Coverage Δ
src/garage/torch/algos/sac.py 98.23% <100.00%> (ø)
src/garage/plotter/plotter.py 59.77% <0.00%> (-3.45%) ⬇️

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 eec1bb3...a38a1fd. Read the comment docs.

@ryanjulian
Copy link
Member

@Mergifyio rebase

@mergify
Copy link
Contributor

mergify bot commented Aug 20, 2020

Command rebase: success

Branch has been successfully rebased

@ahtsan ahtsan force-pushed the avnish-use-max-episode-length-eval branch from ccf66fc to a38a1fd Compare August 20, 2020 17:00
@ryanjulian ryanjulian merged commit cc163fb into master Aug 20, 2020
@ryanjulian ryanjulian deleted the avnish-use-max-episode-length-eval branch August 20, 2020 19:05
@ryanjulian
Copy link
Member

@Mergifyio backport release-2020.06

@mergify
Copy link
Contributor

mergify bot commented Aug 20, 2020

Command backport release-2020.06: failure

No backport have been created

  • Backport to branch release-2020.06 failed

Cherry-pick of cc163fb has failed:

On branch mergify/bp/release-2020.06/pr-1908
Your branch is up to date with 'origin/release-2020.06'.

You are currently cherry-picking commit cc163fbf.
  (fix conflicts and run "git cherry-pick --continue")
  (use "git cherry-pick --abort" to cancel the cherry-pick operation)

Unmerged paths:
  (use "git add <file>..." to mark resolution)

	both modified:   examples/torch/sac_half_cheetah_batch.py
	both modified:   src/garage/torch/algos/sac.py

no changes added to commit (use "git add" and/or "git commit -a")

avnishn pushed a commit that referenced this pull request Sep 3, 2020
Backport #1905, #1975, #1908 to fix problems
with max_eval_path_length being not used by
mtsac and sac, and add checking for incorrect
num_tasks being set in mtsac.
avnishn pushed a commit that referenced this pull request Sep 9, 2020
Backport #1905, #1975, #1908 to fix problems
with max_eval_path_length being not used by
mtsac and sac, and add checking for incorrect
num_tasks being set in mtsac.

Timelimit.truncated modified only when necessary

This issue occurs when there are multiple garage
envs that are nested or timelimit truncated = False
is included in the environment keys.
Previously, our timelimit
truncated logic was written with the idea in mind
that the key was only added when a time limit
truncation occured. If an environment already
has timelimit truncated = False in its keys
then the previous behavior was to set Done = True
which is the incorrect behavior.

That was causing performance degradation
in MTSAC and MTPPO/TRPO.

Now Done is only true in the normal/trivial case,
never if timelimit truncated is False.
avnishn pushed a commit that referenced this pull request Sep 11, 2020
Backport #1905, #1975, #1908 to fix problems
with max_eval_path_length being not used by
mtsac and sac, and add checking for incorrect
num_tasks being set in mtsac.

Timelimit.truncated modified only when necessary

This issue occurs when there are multiple garage
envs that are nested or timelimit truncated = False
is included in the environment keys.
Previously, our timelimit
truncated logic was written with the idea in mind
that the key was only added when a time limit
truncation occured. If an environment already
has timelimit truncated = False in its keys
then the previous behavior was to set Done = True
which is the incorrect behavior.

That was causing performance degradation
in MTSAC and MTPPO/TRPO.

Now Done is only true in the normal/trivial case,
never if timelimit truncated is False.
avnishn pushed a commit that referenced this pull request Sep 11, 2020
Backport #1905, #1975, #1908 to fix problems
with max_eval_path_length being not used by
mtsac and sac, and add checking for incorrect
num_tasks being set in mtsac.

Timelimit.truncated modified only when necessary

This issue occurs when there are multiple garage
envs that are nested or timelimit truncated = False
is included in the environment keys.
Previously, our timelimit
truncated logic was written with the idea in mind
that the key was only added when a time limit
truncation occured. If an environment already
has timelimit truncated = False in its keys
then the previous behavior was to set Done = True
which is the incorrect behavior.

That was causing performance degradation
in MTSAC and MTPPO/TRPO.

Now Done is only true in the normal/trivial case,
never if timelimit truncated is False.
mergify bot pushed a commit that referenced this pull request Sep 11, 2020
Backport #1905, #1975, #1908 to fix problems
with max_eval_path_length being not used by
mtsac and sac, and add checking for incorrect
num_tasks being set in mtsac.

Timelimit.truncated modified only when necessary

This issue occurs when there are multiple garage
envs that are nested or timelimit truncated = False
is included in the environment keys.
Previously, our timelimit
truncated logic was written with the idea in mind
that the key was only added when a time limit
truncation occured. If an environment already
has timelimit truncated = False in its keys
then the previous behavior was to set Done = True
which is the incorrect behavior.

That was causing performance degradation
in MTSAC and MTPPO/TRPO.

Now Done is only true in the normal/trivial case,
never if timelimit truncated is False.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
backport-to-2020.06 Backport this PR to release-2020.06 ready-to-merge
Projects
None yet
Development

Successfully merging this pull request may close these issues.

MTSAC Max Path Length Incorrect
2 participants