ModuleESolver::ESolver_NEP Class Reference

#include <esolver_nep.h>

Inheritance diagram for ModuleESolver::ESolver_NEP:

Collaboration diagram for ModuleESolver::ESolver_NEP:

Public Member Functions
	ESolver_NEP (const std::string &pot_file)
void	before_all_runners (UnitCell &ucell, const Input_para &inp) override
	Initialize the NEP solver with given input parameters and unit cell.
void	runner (UnitCell &ucell, const int istep) override
	Run the NEP solver for a given ion/md step and unit cell.
double	cal_energy () override
	get the total energy without ion kinetic energy
void	cal_force (UnitCell &ucell, ModuleBase::matrix &force) override
	get the computed atomic forces
void	cal_stress (UnitCell &ucell, ModuleBase::matrix &stress) override
	get the computed lattice virials
void	after_all_runners (UnitCell &ucell) override
	Prints the final total energy of the NEP model to the output file.
Public Member Functions inherited from ModuleESolver::ESolver
	ESolver ()
virtual	~ESolver ()
virtual void	others (UnitCell &ucell, const int istep)

Private Member Functions
void	type_map (const UnitCell &ucell)
	determine the type map of NEP model

Private Attributes
std::string	nep_file
	NEP related variables for ESolver_NEP class.
std::vector< int >	atype = {}
	atom type mapping for NEP model
double	nep_potential
	computed potential energy
ModuleBase::matrix	nep_force
	computed atomic forces
ModuleBase::matrix	nep_virial
	computed lattice virials
std::vector< double >	_e
	temporary storage for energy computation
std::vector< double >	_f
	temporary storage for force computation
std::vector< double >	_v
	temporary storage for virial computation

Additional Inherited Members
Public Attributes inherited from ModuleESolver::ESolver
bool	conv_esolver = true
std::string	classname

Constructor & Destructor Documentation

ESolver_NEP()

ModuleESolver::ESolver_NEP::ESolver_NEP ( const std::string &  pot_file )

inline

Member Function Documentation

after_all_runners()

void ESolver_NEP::after_all_runners ( UnitCell &  ucell )

overridevirtual

Prints the final total energy of the NEP model to the output file.

This function prints the final total energy of the NEP model in eV to the output file along with some formatting.

Implements ModuleESolver::ESolver.

before_all_runners()

void ESolver_NEP::before_all_runners ( UnitCell &  ucell, const Input_para &  inp  )

overridevirtual

Initialize the NEP solver with given input parameters and unit cell.

Parameters

inp	input parameters
cell	unitcell information

Implements ModuleESolver::ESolver.

Here is the call graph for this function:

cal_energy()

double ESolver_NEP::cal_energy ( )

overridevirtual

get the total energy without ion kinetic energy

Parameters

etot	the computed energy

Returns

total energy without ion kinetic energy

Implements ModuleESolver::ESolver.

cal_force()

void ESolver_NEP::cal_force ( UnitCell &  ucell, ModuleBase::matrix &  force  )

overridevirtual

get the computed atomic forces

Parameters

force

the computed atomic forces

Implements ModuleESolver::ESolver.

Here is the call graph for this function:

cal_stress()

void ESolver_NEP::cal_stress ( UnitCell &  ucell, ModuleBase::matrix &  stress  )

overridevirtual

get the computed lattice virials

Parameters

stress

the computed lattice virials

Implements ModuleESolver::ESolver.

Here is the call graph for this function:

runner()

void ESolver_NEP::runner ( UnitCell &  ucell, const int  istep  )

overridevirtual

Run the NEP solver for a given ion/md step and unit cell.

Parameters

istep	the current ion/md step
cell	unitcell information

Implements ModuleESolver::ESolver.

Here is the call graph for this function:

type_map()

void ModuleESolver::ESolver_NEP::type_map ( const UnitCell &  ucell )

private

determine the type map of NEP model

Parameters

ucell

unitcell information

Here is the caller graph for this function:

Member Data Documentation

_e

std::vector<double> ModuleESolver::ESolver_NEP::_e

private

temporary storage for energy computation

_f

std::vector<double> ModuleESolver::ESolver_NEP::_f

private

temporary storage for force computation

_v

std::vector<double> ModuleESolver::ESolver_NEP::_v

private

temporary storage for virial computation

atype

std::vector<int> ModuleESolver::ESolver_NEP::atype = {}

private

atom type mapping for NEP model

nep_file

std::string ModuleESolver::ESolver_NEP::nep_file

private

NEP related variables for ESolver_NEP class.

These variables are related to the NEP method and are used in the ESolver_NEP class to compute the potential energy and forces.

Note

These variables are only defined if the __NEP preprocessor macro is defined. directory of NEP model file

nep_force

ModuleBase::matrix ModuleESolver::ESolver_NEP::nep_force

private

computed atomic forces

nep_potential

double ModuleESolver::ESolver_NEP::nep_potential

private

computed potential energy

nep_virial

ModuleBase::matrix ModuleESolver::ESolver_NEP::nep_virial

private

computed lattice virials

---

The documentation for this class was generated from the following files:

• /home/runner/work/abacus-develop/abacus-develop/source/source_esolver/esolver_nep.h
• /home/runner/work/abacus-develop/abacus-develop/source/source_esolver/esolver_nep.cpp