[Ecosystem] Nvidia PhysicsNeMo

[ram-cherukuri]

Contact emails

ramc@nvidia.com

Project summary

Open-source deep-learning framework for building, training, fine-tuning, and inferring Physics AI models using state-of-the-art SciML methods for AI4Science and engineering.

Project description

NVIDIA PhysicsNeMo is an open-source deep-learning framework for building, training, fine-tuning, and inferring Physics AI models using state-of-the-art SciML methods for AI4Science and engineering.

PhysicsNeMo provides Python modules to compose scalable and optimized training and inference pipelines to explore, develop, validate, and deploy AI model architectures tuned for science and engineering to combine physics knowledge with data.

For AI Physics researchers and developers exploring the use of neural operators, GNNs, or transformers, or are interested in Physics-Informed Neural Networks or a hybrid approach in between, PhysicsNeMo provides an optimized stack that will enable them to train their models at scale.

Are there any other projects in the PyTorch Ecosystem similar to yours? If, yes, what are they?

N/A

Project repo URL

https://github.com/NVIDIA/physicsnemo

Additional repos in scope of the application

https://github.com/NVIDIA/physicsnemo-sym
https://github.com/NVIDIA/physicsnemo-curator
https://github.com/NVIDIA/physicsnemo-cfd
https://github.com/NVIDIA/earth2studio

Project license

Apache-2.0 license

GitHub handles of the project maintainer(s)

ktangsali, mnabian, coreyjadams, NickGeneva, loliverhennigh, pzharrington, Alexey-Kamenev, CharlelieLrt, peterdsharpe, dallasfoster, laserkelvin, ram-cherukuri

Is there a corporate or academic entity backing this project? If so, please provide the name and URL of the entity.

Nvidia

Website URL

https://developer.nvidia.com/physicsnemo

Documentation

Getting started: https://docs.nvidia.com/physicsnemo/latest/getting-started/installation.html
User Guide: https://docs.nvidia.com/physicsnemo/latest/user-guide/simple_training_example.html
Example Training Recipes: https://docs.nvidia.com/physicsnemo/latest/examples_catalog.html

How do you build and test the project today (continuous integration)? Please describe.

We have CI system in place for PR pushes - precommit checks followed by pytest. We have release cadence of roughly two months where we build Pip wheels and container.

Version of PyTorch

We keep pace with the latest stable version of PyTorch and constantly upgrade to the latest versions.

Components of PyTorch

The framework is built on top of PyTorch and leverage many components including torch.nn, torch.distributed, torch. cuda etc.

How long do you expect to maintain the project?

Nvidia PhysicsNeMo is intended to be maintained for the long term as it supports a big community of users and has a big group of contributors. Many of the users use the framework for enterprise applications.

Additional information

No response

[lantiga]

Thank you for the submission, we'll get back with the evaluation!

[lantiga]

Thank you @ram-cherukuri for the great contribution. PhysicsNemo is a great project and it's a perfect fit for the ecosystem.
Interestingly other projects entered the ecosystem in the same space, it's certainly a good signal.

Here are my comments. tl;dr every looks great but there's one point related to CI that needs to be discussed.

Functional Requirements:

✅ 1. PyTorch Support: Project must demonstrate benefit to the pytorch community and be validated with the latest versions of PyTorch.
🔴 2. Working CI: Projects must have a demonstrable working CI workflow for all declared architectures

Tests are there, CI development process is there (https://github.com/NVIDIA/physicsnemo/blob/main/CONTRIBUTING.md#continuous-integration-ci) and CI is certainly running somewhere behind the scenes, but it is not publicly accessible. Both the ability to trigger a CI run as well as accessing CI results are not there. This renders community contributions difficult.

This probably reflects the fact that development is happening on a private GitLab repository. Is there a way to at least report success / failures on runs triggered from external PRs on the public repository?

✅ 3. Governance: Projects must have documented technical governance defined and implemented and it must be represented in their Github or Gitlab repository. Example: Review the LF’s Minimum Viable Governance framework.
✅ 4. User experience: The value proposition should be made clear for the project and documented in the README of the project repository.
✅ 5. Quick start: Project should have a working and demonstrable quick start.
✅ 6. Documentation: Projects must have clear and comprehensive documentation.
✅ 7. Permissive licensing: Users must be able to utilize ecosystem projects without licensing concerns. e.g. BSD-3, Apache-2 and MIT licenses
✅ 8. Functional Testing: Users should have developed tests submitted as a part of their project, and be confident that the project will work well with the latest releases of PyTorch.

As a suggestion: there's no clear support matrix or clarification about version support. Since pyproject.toml says "torch>=2.4.0" I'd feature that in the README and docs.
Then of course I'm imagining that YMMV according to what the optional dependencies support.

✅ 9. Packages: Projects need to have support for binary installation options (pip/Conda).
✅ 10. Ongoing maintenance: Project maintainers must be committed to supporting and maintaining their projects for the next 12 months at a minimum. Projects must commit to updating their projects to the latest two releases of PyTorch (e.g. versions 2.6, 2.5.1).

Measurable Requirements:

✅ 1. Core Maintainers: The project must have a minimum of 2 core maintainers
✅ 2. Contributors: The project should have a minimum of 5 total contributors active in the last 90 days.
✅ 3. Commits: The project must have 30 commits in the last 90 days. This is off by like 15 days: it shouldn't be a deal breaker.
✅ 4. Stars: The project must have a minimum of 200 stars on github

Github Layout Requirements:

✅ 1. LICENSE.md at the root of the repository specifying the terms and conditions for using, distributing, and modifying the software. In addition, you should provide information on the license of any third-party code included in the project.
✅ 2. README.md welcomes new community members to the project and explains why the project is useful and how to get started. Should include information about release methodology, cadence, and criteria in the README.
✅ 3. CONTRIBUTING.md explains how to contribute to the project. The file explains the types of contributions needed and how the development process works.
✅ 4. CODEOWNERS that define individuals or teams responsible for code in a repository; document current project owners and Retired committers.
✅ 5. CODE_OF_CONDUCT.md sets the ground rules for participants’ behavior and helps facilitate a friendly, welcoming environment. By default, projects should leverage the [Linux]([https://lfprojects.org/policies/code-of-conduct/) [Foundation](https://lfprojects.org/policies/code-of-conduct/) [Code](https://lfprojects.org/policies/code-of-conduct/) of [Conduct](https://lfprojects.org/policies/code-of-conduct/) unless](https://lfprojects.org/policies/code-of-conduct/)%C2%A0%5B%5BFoundation%5D(https://lfprojects.org/policies/code-of-conduct/)%5D(https://lfprojects.org/policies/code-of-conduct/)%C2%A0%5B%5BCode%5D(https://lfprojects.org/policies/code-of-conduct/)%5D(https://lfprojects.org/policies/code-%5Bof%5D(https://lfprojects.org/policies/code-of-conduct/)-conduct/)%C2%A0%5Bof%5D(https://lfprojects.org/policies/code-of-conduct/)%C2%A0%5B%5BConduct%5D(https://lfprojects.org/policies/code-of-conduct/)%5D(https://lfprojects.org/policies/code-of-conduct/)%C2%A0unless) an alternate Code of Conduct is approved prior.

[ram-cherukuri]

@lantiga Regarding #2 , yes this is due to the current Nvidia internal CI setup. Trying to see if there is a way to report success / failures on runs triggered from external PRs?

[lantiga]

@ram-cherukuri thanks for getting back

Trying to see if there is a way to report success / failures on runs triggered from external PRs

yes external triggers (upon approval, which GH allows to do) + reporting would certainly unblock