Modulus is a PyTorch based deep-learning toolkit for developing deep learning models for physical systems. This package aims to provide useful utilities for physics-constrained and data-driven workflows. Additional information can be found in the Modulus documentation.
The recommended method for installing the latest version of Modulus is using PyPi:
pip install nvidia-modulusThe recommended Modulus docker image can be pulled from the NVIDIA Container Registry:
docker pull nvcr.io/nvidia/modulus/modulus:23.05For a local build of the Modulus Python package from source use:
git clone [email protected]:NVIDIA/modulus.git && cd modulus
pip install --upgrade pip
pip install .To build Modulus docker image:
docker build -t modulus:deploy --target deploy -f Dockerfile .
Alternatively, you can run make container-deploy
To build CI image:
docker build -t modulus:ci --target ci -f Dockerfile .
Alternatively, you can run make container-ci
For guidance on making a contribution to Modulus, see the contributing guidelines.
- Github Discussions: Discuss new architectures, implementations, Physics-ML research, etc.
- GitHub Issues: Bug reports, feature requests, install issues, etc.
- Modulus Forum: The Modulus Forum hosts an audience of new to moderate level users and developers for general chat, online discussions, collaboration, etc.
Modulus is provided under the Apache License 2.0, please see LICENSE.txt for full license text.