Example Notebooks

Example Notebooks#

All notebooks

Example notebooks for aimmd.distributed#

The example notebooks found in the distributed folder (and its subfolders) are all concerned with how to setup, run, and analyze a (large) number of TPS simulations simultaneously, all steered by one common committor model.

Example notebooks for “classic” TPS using openpathsampling#

The example notebooks in the classic folder (and its subfolders) teach you how to use aimmd to perform AI-guided TPS with [openpathsampling]. As opposed to the simulations performed using the aimmd.distributed module, here one committor model steers only one TPS (or other path sampling simulation) and since the actual (T)PS simulation is performed with [openpathsampling] all bonuses and constraints of using it apply. There are currently examples for 2D toy potentials, LiCl dissociation (including solvent), and capped alanine dipeptide.

Example notebooks on learning the transition mechanism from existing data#

The example notebooks in the the apply_on_existing_data (and its subfolders) show how to use aimmd to learn the committor of a transition from existing data without performing any new MD simulations. Currently in here is an example on how to learn the committor of methane hydrate nucleation at different temperatures, then identify the most relevant coordinates, and finally build a low dimensional mathematical expression of the transition using symbolic regression.

Contents