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.

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

[RLlib] Fix action masking example. #47817

Merged
merged 4 commits into from
Sep 25, 2024
Merged

[RLlib] Fix action masking example. #47817

merged 4 commits into from
Sep 25, 2024
+6 −12

Conversation

simonsays1980
Copy link
Collaborator

@simonsays1980 simonsays1980 commented Sep 25, 2024 •
edited
Loading

Why are these changes needed?

Due to changes of the value computation (now in a ConnectorV2) manipulation of the batch as done by the ActionMaskingTorchRLModule the action mask was missing in the batch when _forward_train was entered. This PR fixes this error by adding the action_mask manually during value computation.

Related issue number

Closes #47361

Checks

  • I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR.
  • I've run scripts/format.sh to lint the changes in this PR.
  • I've included any doc changes needed for https://docs.ray.io/en/master/.
    • I've added any new APIs to the API Reference. For example, if I added a
      method in Tune, I've added it in doc/source/tune/api/ under the
      corresponding .rst file.
  • I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
  • Testing Strategy
    • Unit tests
    • Release tests
    • This PR is not tested :(

@simonsays1980
Removed parts of docstrings for LSTMEncoder(s) b/c no linear output l…
58cdc1b 
…ayer is used.

Signed-off-by: simonsays1980 <simon.zehnder@gmail.com>
@simonsays1980
Removed part of docstring from 'RecurrentEncoderConfig' b/c no linear…
eb7fc19 
… layer in recurrent encoders..

Signed-off-by: simonsays1980 <simon.zehnder@gmail.com>
@simonsays1980
Removed parts of 'GRUEncoder(s)' b/c linear output layer is not used.
1c7ede5 
Signed-off-by: simonsays1980 <simon.zehnder@gmail.com>
@simonsays1980
Fixed a bug in 'ActionMaskingTorchRLModule' due to changes in value c…
e6db184 
…omputation.

Signed-off-by: simonsays1980 <simon.zehnder@gmail.com>
@simonsays1980 simonsays1980 added rllib RLlib related issues rllib-models An issue related to RLlib (default or custom) Models. labels Sep 25, 2024
@sven1977 sven1977 changed the title [RLlib] - Fix action masking example [RLlib] Fix action masking example. Sep 25, 2024
@sven1977 sven1977 marked this pull request as ready for review September 25, 2024 17:29
@sven1977 sven1977 self-assigned this Sep 25, 2024
sven1977
sven1977 approved these changes Sep 25, 2024
View reviewed changes
Copy link
Contributor

@sven1977 sven1977 left a comment

Choose a reason for hiding this comment

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

Awesome fix. Thanks @simonsays1980 !!

@sven1977 sven1977 enabled auto-merge (squash) September 25, 2024 17:30
@github-actions github-actions bot added the go add ONLY when ready to merge, run all tests label Sep 25, 2024
@sven1977 sven1977 merged commit 7966130 into ray-project:master Sep 25, 2024
5 of 6 checks passed
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
0b3691a 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
ec07568 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
4a42773 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
cff5b18 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
eaa0426 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
3e7cbd1 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
0d3ad92 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
ce75400 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
ujjawal-khare pushed a commit to ujjawal-khare-27/ray that referenced this pull request Oct 15, 2024
@simonsays1980 @ujjawal-khare
[RLlib] Fix action masking example. (ray-project#47817)
e5c2bd4 
Signed-off-by: ujjawal-khare <ujjawal.khare@dream11.com>
@simonsays1980 simonsays1980 deleted the fix-action-masking-example branch November 22, 2024 10:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@sven1977 sven1977 sven1977 approved these changes

@ArturNiederfahrenhorst ArturNiederfahrenhorst Awaiting requested review from ArturNiederfahrenhorst

Assignees

@sven1977 sven1977

Labels
go add ONLY when ready to merge, run all tests rllib RLlib related issues rllib-models An issue related to RLlib (default or custom) Models. rllib-newstack
Projects
None yet
Milestone
No milestone
2 participants
@simonsays1980 @sven1977