Geoptschnet

class mybigdft.ml_workflows.Geoptschnet(posinp, forcemax=0.01, step_size=0.003, max_iter=300, write_to_disk=False, out_name='')[source]

Bases: object

This class allows to relax the input geometry of a given system in order to find the structure that minimizes the forces. The final result obtained depends on the trained SchnetPack model used.

Parameters
  • posinp (mybigdft.Posinp) – Starting configuration to relax

  • forcemax (float) – Stopping criterion on the forces (in eV/Angstrom)

  • step_size (float) – Step size for each relaxation step

  • max_iter (int) – Maximum number of iterations

  • write_to_disk (bool) – If True, the final positions will be written on disk.

  • out_name (str) – Name of the output file. Default is “final_posinp”

property posinp
Returns

Initial posinp of the geometry optimization procedure

Return type

Posinp

property final_posinp
Returns

Final posinp of the geometry optimization or None if the the optimization has not been completed

Return type

Posinp or None

property forcemax
Returns

Stopping criterion on the forces (in eV/Angstrom)

Return type

float

property step_size
Returns

Step size for each relaxation step

Return type

float

property max_iter
Returns

Maximum number of iterations

Return type

int

property write_to_disk
Returns

If True, the obtained structure will be written on disk.

Return type

bool

property out_name
Returns

Name of the output position file, None if write_to_disk is False.

Return type

str or None

run(model_dir=None, device='cpu', batch_size=128, recenter=False)[source]
Parameters
  • model_dir (str) – Absolute path to the SchnetPack model to use in calculation

  • device (str) – Either ‘cpu’ or ‘cuda’ to run on cpu or gpu

  • batch_size (int) – Size of the mini-batches used in predictions