torch_sim.runners¶

High level runners for atomistic simulations.

This module provides functions for running molecular dynamics simulations and geometry optimizations using various models and integrators. It includes utilities for converting between different atomistic representations and handling simulation state.

Functions

`generate_energy_convergence_fn`	Generate an energy-based convergence function for the convergence_fn argument of the optimize function.
`generate_force_convergence_fn`	Generate a force-based convergence function for the convergence_fn argument of the optimize function.
`integrate`	Simulate a system using a model and integrator.
`optimize`	Optimize a system using a model and optimizer.
`static`	Run single point calculations on a batch of systems.

Classes

`INTEGRATOR_REGISTRY`	dict() -> new empty dictionary dict(mapping) -> new dictionary initialized from a mapping object's (key, value) pairs dict(iterable) -> new dictionary initialized as if via: d = {} for k, v in iterable: d[k] = v dict(**kwargs) -> new dictionary initialized with the name=value pairs in the keyword argument list. For example: dict(one=1, two=2).
`OPTIM_REGISTRY`	dict() -> new empty dictionary dict(mapping) -> new dictionary initialized from a mapping object's (key, value) pairs dict(iterable) -> new dictionary initialized as if via: d = {} for k, v in iterable: d[k] = v dict(**kwargs) -> new dictionary initialized with the name=value pairs in the keyword argument list. For example: dict(one=1, two=2).