2.1.0 rc rebase1#1673
Merged
Merged
Conversation
ktangsali
requested review from
CharlelieLrt,
NickGeneva,
RishikeshRanade,
coreyjadams,
dallasfoster,
loliverhennigh,
mnabian,
peterdsharpe and
pzharrington
as code owners
NickGeneva
approved these changes
May 27, 2026
Collaborator
Author
|
/blossom-ci |
Collaborator
Author
|
Blossom CI passes (see report here) and GitHub CI passed too. |
* add fixes for the nvfuser bug * test(natten): narrow CPU-backward skip to FlexAttention NotImplementedError Cherry-picked test/nn/functional/test_natten.py from upstream commit 7f2451a ("Ci deps group (#1634)"). The previous device == "cpu" early-skip was too broad; this wraps the forward call and only skips on the specific NotImplementedError raised by FlexAttention's CPU-backward guard. If natten picks a different backend (or FlexAttention ever supports CPU backward), the test will run. * black formatting --------- Co-authored-by: Corey adams <6619961+coreyjadams@users.noreply.github.com>
…1642) * add missing dependencies for examples * Replace np.infty with np.inf based on new API * add some more missing dependencies * update use of FusedAdam with native torch's Adam * add tensorboard deps * fix ci issues
* bump up package versions to fix cves * fix greptile comments * update
* fix cve in uv
* few-more-security-fixes
…on3' for improved environment consistency. (#1653)
* Refactor weight initialization to use PyTorch's trunc_normal_ directly - Updated internal weight initialization in distributed AFNO layers and EarthAttention blocks to utilize `torch.nn.init.trunc_normal_` instead of legacy implementations. - Deprecated `trunc_normal_` wrapper in `physicsnemo.nn.module.utils` and removed the in-tree legacy implementation. - Regenerated forward-accuracy reference outputs for several models to align with the new initialization method. - Updated tests to skip on PyTorch versions below 2.12 due to changes in RNG algorithms affecting output consistency. * fix doctest for dit layers --------- Co-authored-by: Kaustubh Tangsali <ktangsali@nvidia.com> Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com>
* Update shard tensor ring attention to use the expected default ring attention results directory * Update examples/minimal/ShardTensorExamples/6_ring_attention/README.md Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Update examples/minimal/ShardTensorExamples/6_ring_attention/benchmark_sharded_attention.py Co-authored-by: Negin Sobhani <negin513@gmail.com> --------- Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Negin Sobhani <negin513@gmail.com>
Prune residual accepted-point conflicts before returning Warp mesh Poisson samples.
…sion (#1658) * Update tensordict dependency constraints and add regression tests for Mesh under torch.compile - Adjusted the tensordict dependency in pyproject.toml to be upper-bounded due to regressions in version 0.12.x, with a note to drop the upper bound once the related PR is merged. - Introduced a new test file for regression testing of the Mesh class to ensure compatibility with torch.compile, specifically addressing issues caused by the tensordict 0.12.x changes. The tests validate that cached properties and data fields behave correctly when compiled. * Update CHANGELOG and bump mlflow and starlette versions - Added a new entry in CHANGELOG detailing the fix for constructing a Mesh inside a torch.compile-traced function, addressing regressions from tensordict 0.12.0. - Updated the mlflow and starlette package versions to 3.12.0 and 0.52.1 respectively, along with their corresponding source distribution and wheel URLs. - Adjusted tensordict dependency constraints to ensure compatibility with the latest changes. * format
* gnn recipes bug fixes * minor fixes
ktangsali
force-pushed
the
2.1.0-rc-rebase1
branch
from
c4386f7 to
bf1a3ba
Compare
Collaborator
Author
|
/blossom-ci |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PhysicsNeMo Pull Request
Description
Rebase RC into main
Checklist
Dependencies
Review Process
All PRs are reviewed by the PhysicsNeMo team before merging.
Depending on which files are changed, GitHub may automatically assign a maintainer for review.
We are also testing AI-based code review tools (e.g., Greptile), which may add automated comments with a confidence score.
This score reflects the AI’s assessment of merge readiness and is not a qualitative judgment of your work, nor is
it an indication that the PR will be accepted / rejected.
AI-generated feedback should be reviewed critically for usefulness.
You are not required to respond to every AI comment, but they are intended to help both authors and reviewers.
Please react to Greptile comments with 👍 or 👎 to provide feedback on their accuracy.