Atomistic applications¶
While metatensor is a generic sparse data container able to store data and metadata for multiple scientific fields, it comes from the field of atomistic machine learning and as such offer some additional facilities for defining and using machine learning models applied to atomistic systems.
The main goal here is to define and train models once, and then be able to re-use them across many different simulation engines (such as LAMMPS, GROMACS, etc.). We strive to achieve this goal without imposing any structure on the model itself, and to allow any model architecture to be used.
This part of metatensor focusses on exporting and importing fully working, already trained models. There are some tools elsewhere to define new models (in the operations and learn submodules).
If you want to train existing architectures with new data or re-use existing trained models, look into the (work in progress!) metatrain project instead.
Why should you use metatensor to define and export your model? What is the point of the interface? How can you use models that follow the interface in your own simulation code?
All of this and more will find answers in this overview!
Learn how to define your own models using metatensor, and how to use these models to run simulation in various simulation engines.
Understand the metadata convention for specific models outputs, such as the potential energy.
Python API referenceRead the documentation for all the classes and functions related to atomistic models in Python.
C++ API referenceRead the documentation for all the classes and functions related to atomistic models in C++.