nnp::Updater Class Reference

Base class for different weight update methods. More...

#include <Updater.h>

Inheritance diagram for nnp::Updater:

Collaboration diagram for nnp::Updater:

Public Member Functions
virtual void	setState (double *state)=0
	Set pointer to current state.
virtual void	setError (double const *const error, std::size_t const size=1)=0
	Set pointer to current error vector.
virtual void	setJacobian (double const *const jacobian, std::size_t const columns=1)=0
	Set pointer to current Jacobi matrix.
virtual void	update ()=0
	Perform single update of state vector.
virtual std::string	status (std::size_t epoch) const =0
	Status report.
virtual std::vector< std::string >	statusHeader () const =0
	Header for status report file.
virtual std::vector< std::string >	info () const =0
	Information about this updater.
virtual void	setupTiming (std::string const &prefix="upd")
	Activate detailed timing.
virtual void	resetTimingLoop ()
	Start a new timing loop (e.g.
virtual std::map< std::string, Stopwatch >	getTiming () const
	Return timings gathered in stopwatch map.

Protected Member Functions
	Updater (std::size_t const sizeState)
	Constructor.

Protected Attributes
bool	timing
	Whether detailed timing is enabled.
bool	timingReset
	Internal loop timer reset switch.
std::size_t	sizeState
	Number of neural network connections (weights + biases).
std::string	prefix
	Prefix for timing stopwatches.
std::map< std::string, Stopwatch >	sw
	Stopwatch map for timing.

Detailed Description

Base class for different weight update methods.

Definition at line 31 of file Updater.h.

Constructor & Destructor Documentation

Updater()

Updater::Updater ( std::size_t const sizeState )

protected

Constructor.

Parameters

[in]

sizeState

Size of the state vector (number of parameters to optimize).

Definition at line 22 of file Updater.cpp.

                                       : timing     (false    ),
                                           timingReset(true     ),
                                           sizeState  (sizeState),
                                           prefix     (""       )
{
}

References prefix, sizeState, timing, and timingReset.

Referenced by nnp::GradientDescent::GradientDescent(), and nnp::KalmanFilter::KalmanFilter().

Here is the caller graph for this function:

Member Function Documentation

setState()

virtual void nnp::Updater::setState ( double * state )

pure virtual

Set pointer to current state.

Parameters

[in,out]

state

Pointer to state vector (weights vector), will be changed in-place upon calling update().

Implemented in nnp::GradientDescent, and nnp::KalmanFilter.

setError()

virtual void nnp::Updater::setError ( double const *const error, std::size_t const size = 1 )

pure virtual

Set pointer to current error vector.

Parameters

[in]	error	Pointer to error (difference between reference and neural network potential output).
[in]	size	Number of error vector entries.

Implemented in nnp::GradientDescent, and nnp::KalmanFilter.

setJacobian()

virtual void nnp::Updater::setJacobian ( double const *const jacobian, std::size_t const columns = 1 )

pure virtual

Set pointer to current Jacobi matrix.

Parameters

[in]	jacobian	Derivatives of error with respect to weights.
[in]	columns	Number of gradients provided.

Note

If there are \(m\) errors and \(n\) weights, the Jacobi matrix is a \(n \times m\) matrix stored in column-major order.

Implemented in nnp::GradientDescent, and nnp::KalmanFilter.

update()

virtual void nnp::Updater::update ( )

pure virtual

Perform single update of state vector.

Implemented in nnp::GradientDescent, and nnp::KalmanFilter.

status()

virtual std::string nnp::Updater::status ( std::size_t epoch ) const

pure virtual

Status report.

Parameters

[in]

epoch

Current epoch.

Returns

Line with current status information.

Implemented in nnp::GradientDescent, and nnp::KalmanFilter.

statusHeader()

virtual std::vector< std::string > nnp::Updater::statusHeader ( ) const

pure virtual

Header for status report file.

Returns

Vector with header lines.

Implemented in nnp::GradientDescent, and nnp::KalmanFilter.

info()

virtual std::vector< std::string > nnp::Updater::info ( ) const

pure virtual

Information about this updater.

Returns

Vector of information lines.

Implemented in nnp::GradientDescent, and nnp::KalmanFilter.

References prefix, and sizeState.

setupTiming()

void Updater::setupTiming ( std::string const & prefix = "upd" )

virtual

Activate detailed timing.

Parameters

[in]

prefix

Prefix used for stopwatch map entries.

Definition at line 29 of file Updater.cpp.

{
    this->prefix = prefix;
    timing = true;
 
    return;
}

References prefix, and timing.

resetTimingLoop()

void Updater::resetTimingLoop ( )

virtual

Start a new timing loop (e.g.

epoch).

Definition at line 42 of file Updater.cpp.

{
    timingReset = false;
    return;
}

References timingReset.

getTiming()

std::map< std::string, Stopwatch > Updater::getTiming ( ) const

virtual

Return timings gathered in stopwatch map.

Returns

Stopwatch map.

Definition at line 37 of file Updater.cpp.

{
    return sw;
}

References sw.

Member Data Documentation

timing

bool nnp::Updater::timing

protected

Whether detailed timing is enabled.

Definition at line 106 of file Updater.h.

Referenced by setupTiming(), and Updater().

timingReset

bool nnp::Updater::timingReset

protected

Internal loop timer reset switch.

Definition at line 108 of file Updater.h.

Referenced by resetTimingLoop(), and Updater().

sizeState

std::size_t nnp::Updater::sizeState

protected

Number of neural network connections (weights + biases).

Definition at line 110 of file Updater.h.

Referenced by nnp::GradientDescent::GradientDescent(), nnp::GradientDescent::info(), nnp::KalmanFilter::info(), info(), nnp::KalmanFilter::KalmanFilter(), nnp::KalmanFilter::setState(), nnp::GradientDescent::status(), nnp::KalmanFilter::status(), nnp::GradientDescent::update(), and Updater().

prefix

std::string nnp::Updater::prefix

protected

Prefix for timing stopwatches.

Definition at line 112 of file Updater.h.

Referenced by info(), setupTiming(), and Updater().

sw

std::map<std::string, Stopwatch> nnp::Updater::sw

protected

Stopwatch map for timing.

Definition at line 114 of file Updater.h.

Referenced by getTiming().

---

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

• /home/runner/work/n2p2/n2p2/src/libnnptrain/Updater.h
• /home/runner/work/n2p2/n2p2/src/libnnptrain/Updater.cpp