NVIDIA / modulus

Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods
https://developer.nvidia.com/modulus
Apache License 2.0
836 stars 190 forks source link

🚀[FEA]: augmentation added for corrdiff training #421

Closed MortezaMardani closed 3 months ago

MortezaMardani commented 3 months ago

Is this a new feature, an improvement, or a change to existing functionality?

New Feature

How would you describe the priority of this feature request

Critical (currently preventing usage)

Please provide a clear description of problem you would like to solve.

currently the augmentation has been deactivated in the modulus codebase for training. it would be very useful to add it. the lag is augment, that refers to augmentation probability. if augment=0, no augmentation.

Describe any alternatives you have considered

No response

mnabian commented 3 months ago

Merged https://github.com/NVIDIA/modulus/pull/426