carcara 25.7.1

Towards Explainable, Scalable, and Accurate Machine-Learned Interatomic Potentials

Carcará

🚧 (Under development) 🚧

Towards Explainable, Scalable, and Accurate Machine-Learned Interatomic Potentials

Installation

From pip

The easiest way to install Carcará is with pip:

pip install carcara

From github

To install Carcará directly from the GitHub repository, run the following commands:

pip install git+https://github.com/seixasgroup/carcara.git

Getting started

Training

model: "MACE"
name: "my_model"

datasets:
  training: "training.xyz"
  validation: "validation.xyz"
  test: "test.xyz"

e3nn_irreps:
  num_channels: 64
  l_max: 1

cutoff_radius: 6.0
message_passing_layers: 2
manybody_correlation: 3

training_attributes:
  energy: "REF_energy"
  forces: "REF_forces"

weights:
  energy: 10
  forces: 1000

seed: 42
device: cpu

Evaluation

# TODO

License

This is an open source code under MIT License.

Acknowledgements

We thank financial support from FAPESP (Grant No. 2022/14549-3), INCT Materials Informatics (Grant No. 406447/2022-5), and CNPq (Grant No. 311324/2020-7).