- Notifications
You must be signed in to change notification settings - Fork 418
Pull requests: pytorch/rl
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[BugFix,Test,Refactor] Refactor tests bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Refactoring Refactoring of an existing feature Tests Incomplete or broken unit tests
#3232 by vmoens was merged Nov 6, 2025 Loading…
[Doc] Huge doc refactoring CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. documentation Improvements or additions to documentation major major refactoring in the code base
#3231 by vmoens was merged Nov 6, 2025 Loading…
[Refactor] Weight sync schemes refactor CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Refactoring Refactoring of an existing feature
#3230 by vmoens was merged Nov 6, 2025 Loading…
[Test] Test RB+Isaac+Ray CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3228 by vmoens was merged Oct 30, 2025 Loading…
[Refactor] Make env creator optional for Ray CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3227 by vmoens was merged Oct 30, 2025 Loading…
[BugFix] Fix schemes and refactor collectors to make them readable bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3226 by vmoens was merged Nov 6, 2025 Loading…
[Refactor] Non-daemonic processes in PEnv bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Refactoring Refactoring of an existing feature
#3225 by vmoens was merged Nov 6, 2025 Loading…
[BugFix] Fix collector devices CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3223 by vmoens was merged Oct 25, 2025 Loading…
[CI] Upgrade doc python version CI Has to do with CI setup (e.g. wheels & builds, tests...) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3222 by vmoens was merged Oct 25, 2025 Loading…
[Refactor] Refactor tool transforms CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. llm/refactor
#3221 by vmoens was closed Oct 26, 2025 Loading…
[Feature] Tool services CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. enhancement New feature or request llm/feature
#3220 by vmoens was merged Nov 6, 2025 Loading…
[Feature] float32 patch CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3219 by vmoens was merged Oct 23, 2025 Loading…
[BugFix] Fix tests bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3218 by vmoens was merged Oct 23, 2025 Loading…
[CI] LLM tests integration CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. llm/ci
#3216 by vmoens was merged Oct 23, 2025 Loading…
[Feature] Composite specs can create named tensors with 'zero' and 'rand' CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. enhancement New feature or request
#3214 by louisfaury was merged Oct 22, 2025 Loading…
3 of 10 tasks
[BugFix] Fix GRPO tests and runs CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3213 by vmoens was merged Oct 22, 2025 Loading…
[CI] Fix benchmarks for LLMs CI Has to do with CI setup (e.g. wheels & builds, tests...) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3212 by vmoens was merged Oct 20, 2025 Loading…
[Quality] Fix flaky test CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. quality code quality
#3211 by vmoens was merged Oct 23, 2025 Loading…
[Tests] Fix vmas seeding test CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Environments Adds or modifies an environment wrapper
#3210 by matteobettini was merged Oct 18, 2025 Loading…
[Feature] Aggregation strategies CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. llm/feature
#3209 by vmoens was merged Nov 6, 2025 Loading…
[Feature] kl_mask_threshold CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. llm/feature
#3208 by vmoens was merged Nov 6, 2025 Loading…
[Feature] CISPO CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. llm/feature Objectives
#3207 by vmoens was merged Nov 6, 2025 Loading…
[Feature] DAPO CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3206 by vmoens was merged Oct 20, 2025 Loading…
[Refactor] Refactor GRPO as a separate class CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3205 by vmoens was merged Oct 20, 2025 Loading…
[Test] Fix flaky parallel test CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3204 by vmoens was merged Oct 18, 2025 Loading…
Previous Next
ProTip! Follow long discussions with comments:>50.