deepmd.pd.model.model.transform_output

Functions

atomic_virial_corr(→ paddle.Tensor)
task_deriv_one(→ tuple[paddle.Tensor, ...)
get_leading_dims(→ list[int])	Get the dimensions of nf x nloc.
take_deriv(→ tuple[paddle.Tensor, paddle.Tensor | None])
fit_output_to_model_output(→ dict[str, paddle.Tensor])	Transform the output of the fitting network to
communicate_extended_output(→ dict[str, paddle.Tensor])	Transform the output of the model network defined on

Module Contents

deepmd.pd.model.model.transform_output.atomic_virial_corr(extended_coord: paddle.Tensor, atom_energy: paddle.Tensor) → paddle.Tensor

deepmd.pd.model.model.transform_output.task_deriv_one(atom_energy: paddle.Tensor, energy: paddle.Tensor, extended_coord: paddle.Tensor, do_virial: bool = True, do_atomic_virial: bool = False, create_graph: bool = True) → tuple[paddle.Tensor, paddle.Tensor | None]

deepmd.pd.model.model.transform_output.get_leading_dims(vv: paddle.Tensor, vdef: deepmd.dpmodel.OutputVariableDef) → list[int]

Get the dimensions of nf x nloc.

deepmd.pd.model.model.transform_output.take_deriv(vv: paddle.Tensor, svv: paddle.Tensor, vdef: deepmd.dpmodel.OutputVariableDef, coord_ext: paddle.Tensor, do_virial: bool = False, do_atomic_virial: bool = False, create_graph: bool = True) → tuple[paddle.Tensor, paddle.Tensor | None]

deepmd.pd.model.model.transform_output.fit_output_to_model_output(fit_ret: dict[str, paddle.Tensor], fit_output_def: deepmd.dpmodel.FittingOutputDef, coord_ext: paddle.Tensor, do_atomic_virial: bool = False, create_graph: bool = True) → dict[str, paddle.Tensor]

Transform the output of the fitting network to the model output.

deepmd.pd.model.model.transform_output.communicate_extended_output(model_ret: dict[str, paddle.Tensor], model_output_def: deepmd.dpmodel.ModelOutputDef, mapping: paddle.Tensor, do_atomic_virial: bool = False) → dict[str, paddle.Tensor]

Transform the output of the model network defined on local and ghost (extended) atoms to local atoms.