deepmd.dpmodel.output_def ========================= .. py:module:: deepmd.dpmodel.output_def Classes ------- .. autoapisummary:: deepmd.dpmodel.output_def.OutputVariableOperation deepmd.dpmodel.output_def.OutputVariableCategory deepmd.dpmodel.output_def.OutputVariableDef deepmd.dpmodel.output_def.FittingOutputDef deepmd.dpmodel.output_def.ModelOutputDef Functions --------- .. autoapisummary:: deepmd.dpmodel.output_def.check_shape deepmd.dpmodel.output_def.check_var deepmd.dpmodel.output_def.model_check_output deepmd.dpmodel.output_def.fitting_check_output deepmd.dpmodel.output_def.get_reduce_name deepmd.dpmodel.output_def.get_deriv_name deepmd.dpmodel.output_def.get_deriv_name_mag deepmd.dpmodel.output_def.get_hessian_name deepmd.dpmodel.output_def.apply_operation deepmd.dpmodel.output_def.check_operation_applied deepmd.dpmodel.output_def.check_deriv deepmd.dpmodel.output_def.do_reduce deepmd.dpmodel.output_def.do_mask deepmd.dpmodel.output_def.do_derivative Module Contents --------------- .. py:function:: check_shape(shape: list[int], def_shape: list[int]) -> None Check if the shape satisfies the defined shape. .. !! processed by numpydoc !! .. py:function:: check_var(var: Any, var_def: Any) -> None .. py:function:: model_check_output(cls: type) -> type 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. .. !! processed by numpydoc !! .. py:function:: fitting_check_output(cls: type) -> type 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. .. !! processed by numpydoc !! .. py:class:: OutputVariableOperation Bases: :py:obj:`enum.IntEnum` Defines the operation of the output variable. .. !! processed by numpydoc !! .. py:attribute:: _NONE :value: 0 No operation. .. !! processed by numpydoc !! .. py:attribute:: REDU :value: 1 Reduce the output variable. .. !! processed by numpydoc !! .. py:attribute:: DERV_R :value: 2 Derivative w.r.t. coordinates. .. !! processed by numpydoc !! .. py:attribute:: DERV_C :value: 4 Derivative w.r.t. cell. .. !! processed by numpydoc !! .. py:attribute:: _SEC_DERV_R :value: 8 Second derivative w.r.t. coordinates. .. !! processed by numpydoc !! .. py:attribute:: MAG :value: 16 Magnetic output. .. !! processed by numpydoc !! .. py:class:: OutputVariableCategory Bases: :py:obj:`enum.IntEnum` Defines the category of the output variable. .. !! processed by numpydoc !! .. py:attribute:: OUT Output variable. (e.g. atom energy) .. !! processed by numpydoc !! .. py:attribute:: REDU Reduced output variable. (e.g. system energy) .. !! processed by numpydoc !! .. py:attribute:: DERV_R Negative derivative w.r.t. coordinates. (e.g. force) .. !! processed by numpydoc !! .. py:attribute:: DERV_C Atomic component of the virial, see PRB 104, 224202 (2021) .. !! processed by numpydoc !! .. py:attribute:: DERV_C_REDU Virial, the transposed negative gradient with cell tensor times cell tensor, see eq 40 JCP 159, 054801 (2023). .. !! processed by numpydoc !! .. py:attribute:: DERV_R_DERV_R Hession matrix, the second derivative w.r.t. coordinates. .. !! processed by numpydoc !! .. py:attribute:: DERV_R_MAG Magnetic part of negative derivative w.r.t. coordinates. (e.g. magnetic force) .. !! processed by numpydoc !! .. py:attribute:: DERV_C_MAG Magnetic part of atomic component of the virial. .. !! processed by numpydoc !! .. py:class:: 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) 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). **atomic** : :ref:`bool ` If the variable is defined for each atom. **category** : :class:`python:int` The category of the output variable. **r_hessian** : :ref:`bool ` If hessian is required **magnetic** : :ref:`bool ` If the derivatives of variable have magnetic parts. **intensive** : :ref:`bool ` It indicates whether the fitting property is intensive or extensive. .. !! processed by numpydoc !! .. py:attribute:: name .. py:attribute:: shape .. py:attribute:: output_size :value: 1 .. py:attribute:: atomic :value: True .. py:attribute:: reducible :value: False .. py:attribute:: r_differentiable :value: False .. py:attribute:: c_differentiable :value: False .. py:attribute:: intensive :value: False .. py:attribute:: category .. py:attribute:: r_hessian :value: False .. py:attribute:: magnetic :value: False .. py:property:: size :type: int .. py:method:: squeeze(dim: int) -> None .. py:class:: FittingOutputDef(var_defs: list[OutputVariableDef]) 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. .. !! processed by numpydoc !! .. py:attribute:: var_defs .. py:method:: __getitem__(key: str) -> OutputVariableDef .. py:method:: get_data() -> dict[str, OutputVariableDef] .. py:method:: keys() .. py:class:: ModelOutputDef(fit_defs: FittingOutputDef) 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 .. !! processed by numpydoc !! .. py:attribute:: def_outp .. py:attribute:: def_redu .. py:attribute:: def_derv_c_redu .. py:attribute:: def_mask .. py:attribute:: var_defs :type: dict[str, OutputVariableDef] .. py:method:: __getitem__(key: str) -> OutputVariableDef .. py:method:: get_data() -> dict[str, OutputVariableDef] .. py:method:: keys() .. py:method:: keys_outp() .. py:method:: keys_redu() .. py:method:: keys_derv_r() .. py:method:: keys_hess_r() .. py:method:: keys_derv_c() .. py:method:: keys_derv_c_redu() .. py:function:: get_reduce_name(name: str) -> str .. py:function:: get_deriv_name(name: str) -> tuple[str, str] .. py:function:: get_deriv_name_mag(name: str) -> tuple[str, str] .. py:function:: get_hessian_name(name: str) -> str .. py:function:: apply_operation(var_def: OutputVariableDef, op: OutputVariableOperation) -> int Apply an operation to the category of a variable definition. :Parameters: **var_def** : :obj:`OutputVariableDef` The variable definition. **op** : :obj:`OutputVariableOperation` The operation to be applied. :Returns: :class:`python:int` The new category of the variable definition. :Raises: :obj:`ValueError` If the operation has been applied to the variable definition, and exceed the maximum limitation. .. !! processed by numpydoc !! .. py:function:: check_operation_applied(var_def: OutputVariableDef, op: OutputVariableOperation) -> bool Check if a operation has been applied to a variable definition. :Parameters: **var_def** : :obj:`OutputVariableDef` The variable definition. **op** : :obj:`OutputVariableOperation` The operation to be checked. :Returns: :ref:`bool ` True if the operation has been applied, False otherwise. .. !! processed by numpydoc !! .. py:function:: check_deriv(var_def: OutputVariableDef) -> bool Check if a variable is obtained by derivative. .. !! processed by numpydoc !! .. py:function:: do_reduce(def_outp_data: dict[str, OutputVariableDef]) -> dict[str, OutputVariableDef] .. py:function:: do_mask(def_outp_data: dict[str, OutputVariableDef]) -> dict[str, OutputVariableDef] .. py:function:: do_derivative(def_outp_data: dict[str, OutputVariableDef]) -> tuple[dict[str, OutputVariableDef], dict[str, OutputVariableDef]]