AeroGraphNet example

[Alexey-Kamenev]

Modulus Pull Request

Description

This example demonstrates how to train MeshGraphNet model on two automotive aerodynamics datasets, Ahmed Body and DrivAerNet.

The example uses Hydra experiments as a way to specify experiment configuration. For example, to train default MGN on DrivAerNet dataset, the following command can be used:

python train.py +experiment=drivaernet/mgn data.data_dir=/data/DrivAerNet/

See the example README for more information.

The example also contains visualization of the meshes which allows to better interpret the training progress.

Checklist

• [x] I am familiar with the Contributing Guidelines.
• [x] New or existing tests cover these changes.
• [x] The documentation is up to date with these changes.
• [x] The CHANGELOG.md is up to date with these changes.
• [x] An issue is linked to this pull request.

Dependencies

[Alexey-Kamenev]

/blossom-ci

[Alexey-Kamenev]

/blossom-ci

[Alexey-Kamenev]

/blossom-ci

[mnabian]

This PR looks great! Left a few minor comments.

[Alexey-Kamenev]

/blossom-ci

[Alexey-Kamenev]

/blossom-ci