---

_gnn_trainer Struct Reference

#include <gnn_trainer.h>

Collaboration diagram for _gnn_trainer:

[legend]

Data Fields
const char *	type
gnn_node *	node
gnn_criterion *	crit
gnn_dataset *	data
size_t	n
size_t	s
size_t	epoch
double	sume
double	sump
gnn_grad *	grad
gnn_trainer_reset_type	reset
gnn_trainer_train_type	train
gnn_trainer_destroy_type	destroy

---

Field Documentation

gnn_criterion* _gnn_trainer::crit

A pointer to the criterion. Definition at line 100 of file gnn_trainer.h.

gnn_dataset* _gnn_trainer::data

A pointer to the dataset. Definition at line 101 of file gnn_trainer.h.

gnn_trainer_destroy_type _gnn_trainer::destroy

The "destroy" function pointer. Definition at line 113 of file gnn_trainer.h.

size_t _gnn_trainer::epoch

The number of the current epoch. Definition at line 105 of file gnn_trainer.h.

gnn_grad* _gnn_trainer::grad

The internal batch gradients evaluation buffers. Definition at line 109 of file gnn_trainer.h.

size_t _gnn_trainer::n

The size of the batches to be processed. Definition at line 103 of file gnn_trainer.h.

gnn_node* _gnn_trainer::node

A pointer to the node to be trained. Definition at line 99 of file gnn_trainer.h.

gnn_trainer_reset_type _gnn_trainer::reset

The "reset" function pointer. Definition at line 111 of file gnn_trainer.h.

size_t _gnn_trainer::s

The number of the next pattern to be evaluated. Definition at line 104 of file gnn_trainer.h.

double _gnn_trainer::sume

The sum of the epochs evaluated criterions. Definition at line 106 of file gnn_trainer.h.

double _gnn_trainer::sump

The sum of the patterns' weights viewed so far. Definition at line 107 of file gnn_trainer.h.

gnn_trainer_train_type _gnn_trainer::train

The "train" function pointer. Definition at line 112 of file gnn_trainer.h.

const char* _gnn_trainer::type

The type of the trainer. Definition at line 97 of file gnn_trainer.h.

---

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

• gnn_trainer.h

---