nest::WeightOptimizerGradientDescent Class Reference

Base class implementing a gradient descent weight optimizer model. More...

#include <weight_optimizer.h>

Inheritance diagram for nest::WeightOptimizerGradientDescent:

Collaboration diagram for nest::WeightOptimizerGradientDescent:

Public Member Functions
	WeightOptimizerGradientDescent ()
	Default constructor.
	WeightOptimizerGradientDescent (const WeightOptimizerGradientDescent &)=default
	Copy constructor.
WeightOptimizerGradientDescent &	operator= (const WeightOptimizerGradientDescent &)=delete
	Assignment operator.
Public Member Functions inherited from nest::WeightOptimizer
	WeightOptimizer ()
	Default constructor.
virtual	~WeightOptimizer ()
	Destructor.
	WeightOptimizer (const WeightOptimizer &)=default
	Copy constructor.
WeightOptimizer &	operator= (const WeightOptimizer &)=delete
	Assignment operator.
virtual void	get_status (Dictionary &d) const
	Get parameter dictionary.
virtual void	set_status (const Dictionary &d)
	Update values in parameter dictionary.
double	optimized_weight (const WeightOptimizerCommonProperties &cp, const size_t idx_current_update, const double gradient, double weight)
	Return optimized weight based on current weight.

Private Member Functions
double	optimize_ (const WeightOptimizerCommonProperties &cp, double weight, size_t current_opt_step) override
	Perform specific optimization.

Additional Inherited Members
Protected Attributes inherited from nest::WeightOptimizer
double	cumulative_gradient_
	Cumulative gradient over the current batch.
size_t	optimization_step_
	Current optimization step, whereby optimization happens every batch_size_ steps.
double	eta_current_
	Synapse-specific learning rate when the history for its upcoming weight update was archived.
long	n_optimize_
	Number of optimizations.

Detailed Description

Base class implementing a gradient descent weight optimizer model.

Constructor & Destructor Documentation

WeightOptimizerGradientDescent() [1/2]

nest::WeightOptimizerGradientDescent::WeightOptimizerGradientDescent

(

)

Default constructor.

WeightOptimizerGradientDescent() [2/2]

nest::WeightOptimizerGradientDescent::WeightOptimizerGradientDescent ( const WeightOptimizerGradientDescent &  )

default

Copy constructor.

Member Function Documentation

operator=()

WeightOptimizerGradientDescent & nest::WeightOptimizerGradientDescent::operator= ( const WeightOptimizerGradientDescent &  )

delete

Assignment operator.

optimize_()

double nest::WeightOptimizerGradientDescent::optimize_ ( const WeightOptimizerCommonProperties &  cp, double  weight, size_t  current_opt_step  )

overrideprivatevirtual

Perform specific optimization.

Implements nest::WeightOptimizer.

References nest::WeightOptimizer::cumulative_gradient_, and nest::WeightOptimizer::eta_current_.

---

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

• weight_optimizer.h
• weight_optimizer.cpp