deepmd.dpmodel

deepmd.dpmodel#

Submodules#

Attributes#

`DEFAULT_PRECISION`
`PRECISION_DICT`

Classes#

`NativeOP`	The unit operation of a native model.
`DPModelCommon`
`FittingOutputDef`	Defines the shapes and other properties of the fitting network outputs.
`ModelOutputDef`	Defines the shapes and other properties of the model outputs.
`OutputVariableDef`	Defines the shape and other properties of the one output variable.

Functions#

`fitting_check_output`(cls)	Check if the output of the Fitting is consistent with the definition.
`get_deriv_name`(→ tuple[str, str])
`get_hessian_name`(→ str)
`get_reduce_name`(→ str)
`model_check_output`(cls)	Check if the output of the Model is consistent with the definition.

Package Contents#

deepmd.dpmodel.DEFAULT_PRECISION = 'float64'[source]#

deepmd.dpmodel.PRECISION_DICT[source]#

class deepmd.dpmodel.NativeOP[source]#

Bases: abc.ABC

The unit operation of a native model.

abstract call(*args, **kwargs)[source]#: Forward pass in NumPy implementation.

__call__(*args, **kwargs)[source]#: Forward pass in NumPy implementation.

class deepmd.dpmodel.DPModelCommon[source]#

classmethod update_sel(train_data: deepmd.utils.data_system.DeepmdDataSystem, type_map: list[str] | None, local_jdata: dict) → tuple[dict, float | None][source]#

Update the selection and perform neighbor statistics.

Parameters:

train_dataDeepmdDataSystem: data used to do neighbor statistics
type_maplist[str], optional: The name of each type of atoms
local_jdatadict: The local data refer to the current class

Returns:

dict: The updated local data
float: The minimum distance between two atoms

class deepmd.dpmodel.FittingOutputDef(var_defs: list[OutputVariableDef])[source]#

Defines the shapes and other properties of the fitting network outputs.

It is assume that the fitting network output variables for each local atom. This class defines all the outputs.

Parameters:

var_defs: List of output variable definitions.

var_defs#

__getitem__(key: str) → OutputVariableDef[source]#

get_data() → dict[str, OutputVariableDef][source]#

keys()[source]#

class deepmd.dpmodel.ModelOutputDef(fit_defs: FittingOutputDef)[source]#

Defines the shapes and other properties of the model outputs.

The model reduce and differentiate fitting outputs if applicable. If a variable is named by foo, then the reduced variable is called foo_redu, the derivative w.r.t. coordinates is called foo_derv_r and the derivative w.r.t. cell is called foo_derv_c.

Parameters:

fit_defs: Definition for the fitting net output

def_outp#

def_redu#

def_derv_c_redu#

def_mask#

var_defs: dict[str, OutputVariableDef]#

__getitem__(key: str) → OutputVariableDef[source]#

get_data() → dict[str, OutputVariableDef][source]#

keys()[source]#

keys_outp()[source]#

keys_redu()[source]#

keys_derv_r()[source]#

keys_hess_r()[source]#

keys_derv_c()[source]#

keys_derv_c_redu()[source]#

class deepmd.dpmodel.OutputVariableDef(name: str, shape: list[int], reducible: bool = False, r_differentiable: bool = False, c_differentiable: bool = False, atomic: bool = True, category: int = OutputVariableCategory.OUT.value, r_hessian: bool = False, magnetic: bool = False, intensive: bool = False)[source]#

Defines the shape and other properties of the one output variable.

It is assume that the fitting network output variables for each local atom. This class defines one output variable, including its name, shape, reducibility and differentiability.

Parameters:

name: Name of the output variable. Notice that the xxxx_redu, xxxx_derv_c, xxxx_derv_r are reserved names that should not be used to define variables.
shape: The shape of the variable. e.g. energy should be [1], dipole should be [3], polarizabilty should be [3,3].
reducible: If the variable is reduced.
r_differentiable: If the variable is differentiated with respect to coordinates of atoms. Only reducible variable are differentiable. Negative derivative w.r.t. coordinates will be calculated. (e.g. force)
c_differentiable: If the variable is differentiated with respect to the cell tensor (pbc case). Only reducible variable are differentiable. Virial, the transposed negative gradient with cell tensor times cell tensor, will be calculated, see eq 40 JCP 159, 054801 (2023).
atomicbool: If the variable is defined for each atom.
categoryint: The category of the output variable.
r_hessianbool: If hessian is required
magneticbool: If the derivatives of variable have magnetic parts.
intensivebool: It indicates whether the fitting property is intensive or extensive.

name#

shape#

output_size = 1#

atomic#

reducible#

r_differentiable#

c_differentiable#

intensive#

category#

r_hessian#

magnetic#

property size#

squeeze(dim) → None[source]#

deepmd.dpmodel.fitting_check_output(cls)[source]#

Check if the output of the Fitting is consistent with the definition.

Two methods are assumed to be provided by the Fitting: 1. Fitting.output_def that gives the output definition. 2. Fitting.__call__ defines the forward path of the fitting.

deepmd.dpmodel.get_deriv_name(name: str) → tuple[str, str][source]#

deepmd.dpmodel.get_hessian_name(name: str) → str[source]#

deepmd.dpmodel.get_reduce_name(name: str) → str[source]#

deepmd.dpmodel.model_check_output(cls)[source]#

Check if the output of the Model is consistent with the definition.

Two methods are assumed to be provided by the Model: 1. Model.output_def that gives the output definition. 2. Model.__call__ that defines the forward path of the model.