xmm::TrainingEvent Class Reference

Event for monitoring the training process. More...

#include <xmmModelSingleClass.hpp>

Public Types
enum	Status {   Run, Error, Cancel, Done,   Alldone }
	Status of the training process. More...
enum	ErrorType { ErrorType::ConvergenceError }
	Type of Training Error. More...

Public Member Functions
	TrainingEvent (void *model_, std::string label_, Status status_=Status::Run)
	Constructor. More...
	TrainingEvent (TrainingEvent const &src)
	Copy constructor. More...
TrainingEvent &	operator= (TrainingEvent const &src)
	Assignment operator. More...

Public Attributes
void *	model
	Source Model. More...
std::string	label
	Label of the source model, if any. More...
Status	status
	Status of the training process. More...
float	progression
	progression within the training algorithm More...
double	log_likelihood
	Log-likelihood of the data given the model's parameters at the en of training. More...
double	iterations
	Number of EM iterations. More...

Detailed Description

Event for monitoring the training process.

Member Enumeration Documentation

enum xmm::TrainingEvent::ErrorType

strong

Type of Training Error.

Enumerator
ConvergenceError	Convergence Errors (numerical errors in the training process). these errors can be due to issues with the training data (e.g. nan values), or bad choice of parameters (e.g. regularization too low).

enum xmm::TrainingEvent::Status

Status of the training process.

Enumerator
Run	Training is still running.
Error	An error occured during training.
Cancel	The training has been cancelled.
Done	Training is done without error.
Alldone	The training of all classes has finished.

Constructor & Destructor Documentation

xmm::TrainingEvent::TrainingEvent ( void *  model_, std::string  label_, Status  status_ = Status::Run  )

inline

Constructor.

Parameters

model_	Source model
label_	Label of the model sending the notification
status_	status of the training process

xmm::TrainingEvent::TrainingEvent ( TrainingEvent const &  src )

inline

Copy constructor.

Parameters

src	source event

Member Function Documentation

TrainingEvent& xmm::TrainingEvent::operator= ( TrainingEvent const &  src )

inline

Assignment operator.

Parameters

src	source event

Member Data Documentation

double xmm::TrainingEvent::iterations

Number of EM iterations.

std::string xmm::TrainingEvent::label

Label of the source model, if any.

double xmm::TrainingEvent::log_likelihood

Log-likelihood of the data given the model's parameters at the en of training.

void* xmm::TrainingEvent::model

Source Model.

float xmm::TrainingEvent::progression

progression within the training algorithm

Status xmm::TrainingEvent::status

Status of the training process.

---

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

• xmmModelSingleClass.hpp