Thrill  0.1
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
StochasticGradientDescent< Vector > Class Template Reference

Detailed Description

template<typename Vector>
class examples::stochastic_gradient_descent::StochasticGradientDescent< Vector >

Definition at line 99 of file stochastic_gradient_descent.hpp.

#include <stochastic_gradient_descent.hpp>

Public Member Functions

 StochasticGradientDescent (size_t num_iterations, double mini_batch_fraction, double step_size, double tolerance)
 
Vector optimize (const DIA< DataPoint< Vector > > &input_points, const Vector &initial_weights)
 do the actual computation More...
 

Private Member Functions

bool is_converged (Vector &old, Vector &curr, double tolerance)
 

Private Attributes

double mini_batch_fraction
 
size_t num_iterations
 
double step_size
 
double tolerance
 

Constructor & Destructor Documentation

StochasticGradientDescent ( size_t  num_iterations,
double  mini_batch_fraction,
double  step_size,
double  tolerance 
)
inline

Definition at line 102 of file stochastic_gradient_descent.hpp.

Member Function Documentation

bool is_converged ( Vector old,
Vector curr,
double  tolerance 
)
inlineprivate

Member Data Documentation

double mini_batch_fraction
private
size_t num_iterations
private
double step_size
private
double tolerance
private

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