这是indexloc提供的服务，不要输入任何密码
Skip to content

Fix SAC loss explode #333

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 10 commits into from
Apr 4, 2021
Merged

Fix SAC loss explode #333

merged 10 commits into from
Apr 4, 2021

Conversation

ChenDRAG
Copy link
Collaborator

@ChenDRAG ChenDRAG commented Apr 1, 2021 •
edited
Loading

fix #332

@ChenDRAG ChenDRAG mentioned this pull request Apr 1, 2021
Closed
8 tasks
Trinkle23897
Trinkle23897 reviewed Apr 1, 2021
View reviewed changes
@Trinkle23897 Trinkle23897 changed the title Fix sac Fix SAC loss explode Apr 1, 2021
@codecov-io
Copy link

codecov-io commented Apr 1, 2021 •
edited
Loading

Codecov Report

Merging #333 (0bb25ee) into master (825da9b) will increase coverage by 0.30%.
The diff coverage is 87.50%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master     #333      +/-   ##
==========================================
+ Coverage   93.88%   94.19%   +0.30%     
==========================================
  Files          51       51              
  Lines        3256     3256              
==========================================
+ Hits         3057     3067      +10     
+ Misses        199      189      -10
Flag Coverage Δ
unittests 94.19% <87.50%> (+0.30%) ⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files Coverage Δ
tianshou/policy/base.py 78.62% <ø> (-1.53%) ⬇️
tianshou/policy/modelbased/psrl.py 97.40% <ø> (ø)
tianshou/policy/modelfree/td3.py 100.00% <ø> (ø)
tianshou/policy/modelfree/sac.py 98.82% <83.33%> (+13.93%) ⬆️
tianshou/policy/__init__.py 100.00% <100.00%> (ø)
tianshou/policy/modelfree/ddpg.py 96.29% <100.00%> (+0.04%) ⬆️

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 825da9b...0bb25ee. Read the comment docs.

Trinkle23897
Trinkle23897 reviewed Apr 1, 2021
View reviewed changes
@Trinkle23897 Trinkle23897 requested a review from danagi April 2, 2021 06:01
@ChenDRAG
Copy link
Collaborator Author

ChenDRAG commented Apr 2, 2021

Could you give some opinion, please? @danagi

danagi
danagi previously approved these changes Apr 3, 2021
View reviewed changes
@Trinkle23897 Trinkle23897 merged commit dd4a011 into thu-ml:master Apr 4, 2021
BFAnas pushed a commit to BFAnas/tianshou that referenced this pull request May 5, 2024
@ChenDRAG
Fix SAC loss explode (thu-ml#333)
23c77ad 
* change SAC action_bound_method to "clip" (tanh is hardcoded in forward)

* docstring update

* modelbase -> modelbased
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@Trinkle23897 Trinkle23897 Trinkle23897 approved these changes

@danagi danagi danagi left review comments

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

SAC's loss explode on Hopper-v3
4 participants
@ChenDRAG @codecov-io @Trinkle23897 @danagi