Multi-layer perceptron neural network condition functions. More...

#include "cond_neural.h"
#include "neural_activations.h"
#include "neural_layer_connected.h"
#include "neural_layer_convolutional.h"
#include "neural_layer_dropout.h"
#include "neural_layer_lstm.h"
#include "neural_layer_maxpool.h"
#include "neural_layer_noise.h"
#include "neural_layer_recurrent.h"
#include "neural_layer_softmax.h"
#include "utils.h"

Include dependency graph for cond_neural.c:

Functions
void	cond_neural_init (const struct XCSF xcsf, struct Cl c)
	Creates and initialises a neural network condition. More...

void	cond_neural_free (const struct XCSF xcsf, const struct Cl c)
	Frees the memory used by a neural network condition. More...

void	cond_neural_copy (const struct XCSF xcsf, struct Cl dest, const struct Cl *src)
	Copies a neural network condition from one classifier to another. More...

void	cond_neural_cover (const struct XCSF xcsf, const struct Cl c, const double *x)
	Generates a neural network that matches the current input. More...

void	cond_neural_update (const struct XCSF xcsf, const struct Cl c, const double x, const double y)
	Dummy update function. More...

bool	cond_neural_match (const struct XCSF xcsf, const struct Cl c, const double *x)
	Generates a neural network that matches the current input. More...

bool	cond_neural_mutate (const struct XCSF xcsf, const struct Cl c)
	Mutates a neural network condition with the self-adaptive rates. More...

bool	cond_neural_crossover (const struct XCSF xcsf, const struct Cl c1, const struct Cl *c2)
	Dummy crossover function. More...

bool	cond_neural_general (const struct XCSF xcsf, const struct Cl c1, const struct Cl *c2)
	Dummy general function. More...

void	cond_neural_print (const struct XCSF xcsf, const struct Cl c)
	Prints a neural network condition. More...

double	cond_neural_size (const struct XCSF xcsf, const struct Cl c)
	Returns the size of a neural network condition. More...

size_t	cond_neural_save (const struct XCSF xcsf, const struct Cl c, FILE *fp)
	Writes a neural network condition to a file. More...

size_t	cond_neural_load (const struct XCSF xcsf, struct Cl c, FILE *fp)
	Reads a neural network condition from a file. More...

int	cond_neural_neurons (const struct XCSF xcsf, const struct Cl c, int layer)
	Returns the number of neurons in a neural condition layer. More...

int	cond_neural_connections (const struct XCSF xcsf, const struct Cl c, int layer)
	Returns the number of active connections in a neural condition layer. More...

int	cond_neural_layers (const struct XCSF xcsf, const struct Cl c)
	Returns the number of layers within a neural network condition. More...

char *	cond_neural_json_export (const struct XCSF xcsf, const struct Cl c)
	Returns a json formatted string representation of a neural condition. More...

void	cond_neural_json_import (const struct XCSF xcsf, struct Cl c, const cJSON *json)
	Creates a neural condition from a cJSON object. More...

char *	cond_neural_param_json_import (struct XCSF xcsf, cJSON json)
	Sets the neural network parameters from a cJSON object. More...

void	cond_neural_param_defaults (struct XCSF *xcsf)
	Initialises default neural condition parameters. More...

Detailed Description

Multi-layer perceptron neural network condition functions.

Author: Richard Preen rpree.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om

Copyright: The Authors.

Date: 2016–2022.

Definition in file cond_neural.c.

Function Documentation

◆ cond_neural_connections()

int cond_neural_connections	(	const struct XCSF *	xcsf,
		const struct Cl *	c,
		int	layer
	)

Returns the number of active connections in a neural condition layer.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier maintaining a neural network condition.
[in]	layer	Position of a layer in the network.

Returns: The current number of active connections in a layer.

Definition at line 273 of file cond_neural.c.

References Cl::cond, Llist::layer, Layer::n_active, CondNeural::net, Llist::prev, and Net::tail.

Referenced by clset_mean_cond_connections().

Here is the caller graph for this function:

◆ cond_neural_copy()

void cond_neural_copy	(	const struct XCSF *	xcsf,
		struct Cl *	dest,
		const struct Cl *	src
	)

Copies a neural network condition from one classifier to another.

Parameters

[in]	xcsf	XCSF data structure.
[in]	dest	Destination classifier.
[in]	src	Source classifier.

Definition at line 70 of file cond_neural.c.

References Cl::cond, CondNeural::net, and neural_copy().

Here is the call graph for this function:

◆ cond_neural_cover()

void cond_neural_cover	(	const struct XCSF *	xcsf,
		const struct Cl *	c,
		const double *	x
	)

Generates a neural network that matches the current input.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is being covered.
[in]	x	Input state to cover.

Definition at line 86 of file cond_neural.c.

References Cl::cond, cond_neural_match(), CondNeural::net, and neural_rand().

Here is the call graph for this function:

◆ cond_neural_crossover()

bool cond_neural_crossover	(	const struct XCSF *	xcsf,
		const struct Cl *	c1,
		const struct Cl *	c2
	)

Dummy crossover function.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c1	First classifier whose condition is being crossed.
[in]	c2	Second classifier whose condition is being crossed.

Returns: False

Definition at line 150 of file cond_neural.c.

◆ cond_neural_free()

void cond_neural_free	(	const struct XCSF *	xcsf,
		const struct Cl *	c
	)

Frees the memory used by a neural network condition.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is to be freed.

Definition at line 55 of file cond_neural.c.

References Cl::cond, CondNeural::net, and neural_free().

Here is the call graph for this function:

◆ cond_neural_general()

bool cond_neural_general	(	const struct XCSF *	xcsf,
		const struct Cl *	c1,
		const struct Cl *	c2
	)

Dummy general function.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c1	Classifier whose condition is tested to be more general.
[in]	c2	Classifier whose condition is tested to be more specific.

Returns: False.

Definition at line 167 of file cond_neural.c.

◆ cond_neural_init()

void cond_neural_init	(	const struct XCSF *	xcsf,
		struct Cl *	c
	)

Creates and initialises a neural network condition.

Parameters

[in]	xcsf	The XCSF data structure.
[in]	c	The classifier whose condition is to be initialised.

Definition at line 42 of file cond_neural.c.

References Cl::cond, and neural_create().

Here is the call graph for this function:

◆ cond_neural_json_export()

char* cond_neural_json_export	(	const struct XCSF *	xcsf,
		const struct Cl *	c
	)

Returns a json formatted string representation of a neural condition.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is to be returned.

Returns: String encoded in json format.

Definition at line 312 of file cond_neural.c.

References Cl::cond, CondNeural::net, and neural_json_export().

Referenced by cond_neural_print().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ cond_neural_json_import()

void cond_neural_json_import	(	const struct XCSF *	xcsf,
		struct Cl *	c,
		const cJSON *	json
	)

Creates a neural condition from a cJSON object.

Parameters

[in]	xcsf	The XCSF data structure.
[in,out]	c	The classifier to initialise.
[in]	json	cJSON object.

Definition at line 334 of file cond_neural.c.

References Cl::cond, CondNeural::net, and neural_json_import().

Here is the call graph for this function:

◆ cond_neural_layers()

int cond_neural_layers	(	const struct XCSF *	xcsf,
		const struct Cl *	c
	)

Returns the number of layers within a neural network condition.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier maintaining a neural network condition.

Returns: The number of layers.

Definition at line 297 of file cond_neural.c.

References Cl::cond, Net::n_layers, and CondNeural::net.

Referenced by clset_mean_cond_layers().

Here is the caller graph for this function:

◆ cond_neural_load()

size_t cond_neural_load	(	const struct XCSF *	xcsf,
		struct Cl *	c,
		FILE *	fp
	)

Reads a neural network condition from a file.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is to be read.
[in]	fp	Pointer to the file to be read.

Returns: The number of elements read.

Definition at line 228 of file cond_neural.c.

References Cl::cond, and neural_load().

Here is the call graph for this function:

◆ cond_neural_match()

bool cond_neural_match	(	const struct XCSF *	xcsf,
		const struct Cl *	c,
		const double *	x
	)

Generates a neural network that matches the current input.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is being covered.
[in]	x	Input state to cover.

Definition at line 118 of file cond_neural.c.

References Cl::cond, CondNeural::net, neural_output(), and neural_propagate().

Referenced by cond_neural_cover().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ cond_neural_mutate()

bool cond_neural_mutate	(	const struct XCSF *	xcsf,
		const struct Cl *	c
	)

Mutates a neural network condition with the self-adaptive rates.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is being mutated.

Returns: Whether any alterations were made.

Definition at line 135 of file cond_neural.c.

References Cl::cond, CondNeural::net, and neural_mutate().

Here is the call graph for this function:

◆ cond_neural_neurons()

int cond_neural_neurons	(	const struct XCSF *	xcsf,
		const struct Cl *	c,
		int	layer
	)

Returns the number of neurons in a neural condition layer.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier maintaining a neural network prediction.
[in]	layer	Position of a layer in the network.

Returns: The current number of neurons in a layer.

Definition at line 245 of file cond_neural.c.

References Cl::cond, CONVOLUTIONAL, Llist::layer, Layer::n_filters, Layer::n_outputs, CondNeural::net, Llist::prev, Net::tail, and Layer::type.

Referenced by clset_mean_cond_neurons().

Here is the caller graph for this function:

◆ cond_neural_param_defaults()

void cond_neural_param_defaults ( struct XCSF * xcsf )

Initialises default neural condition parameters.

Parameters

[in] xcsf The XCSF data structure.

Definition at line 383 of file cond_neural.c.

References CONNECTED, ArgsLayer::evolve_connect, ArgsLayer::evolve_neurons, ArgsLayer::evolve_weights, ArgsLayer::function, layer_args_copy(), layer_args_init(), LINEAR, LOGISTIC, ArgsLayer::max_neuron_grow, ArgsLayer::n_init, ArgsLayer::n_inputs, ArgsLayer::n_max, ArgsLayer::next, and ArgsLayer::type.

Referenced by cond_param_defaults().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ cond_neural_param_json_import()

char* cond_neural_param_json_import	(	struct XCSF *	xcsf,
		cJSON *	json
	)

Sets the neural network parameters from a cJSON object.

Parameters

[in,out]	xcsf	The XCSF data structure.
[in]	json	cJSON object.

Returns: NULL if successful; or the name of parameter if not found.

Definition at line 353 of file cond_neural.c.

References layer_args_free(), layer_args_init(), layer_args_json_import(), layer_args_validate(), ArgsLayer::n_inputs, and ArgsLayer::next.

Referenced by cond_param_json_import().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ cond_neural_print()

void cond_neural_print	(	const struct XCSF *	xcsf,
		const struct Cl *	c
	)

Prints a neural network condition.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is to be printed.

Definition at line 182 of file cond_neural.c.

References cond_neural_json_export().

Here is the call graph for this function:

◆ cond_neural_save()

size_t cond_neural_save	(	const struct XCSF *	xcsf,
		const struct Cl *	c,
		FILE *	fp
	)

Writes a neural network condition to a file.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is to be written.
[in]	fp	Pointer to the file to be written.

Returns: The number of elements written.

Definition at line 212 of file cond_neural.c.

References Cl::cond, CondNeural::net, and neural_save().

Here is the call graph for this function:

◆ cond_neural_size()

double cond_neural_size	(	const struct XCSF *	xcsf,
		const struct Cl *	c
	)

Returns the size of a neural network condition.

See also: neural_size()

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition size to return.

Returns: The network size.

Definition at line 197 of file cond_neural.c.

References Cl::cond, CondNeural::net, and neural_size().

Here is the call graph for this function:

◆ cond_neural_update()

void cond_neural_update	(	const struct XCSF *	xcsf,
		const struct Cl *	c,
		const double *	x,
		const double *	y
	)

Dummy update function.

Parameters

[in]	xcsf	XCSF data structure.
[in]	c	Classifier whose condition is to be updated.
[in]	x	Input state.
[in]	y	Truth/payoff value.

Definition at line 102 of file cond_neural.c.

Functions

Detailed Description

Function Documentation

◆ cond_neural_connections()

◆ cond_neural_copy()

◆ cond_neural_cover()

◆ cond_neural_crossover()

◆ cond_neural_free()

◆ cond_neural_general()

◆ cond_neural_init()

◆ cond_neural_json_export()

◆ cond_neural_json_import()

◆ cond_neural_layers()

◆ cond_neural_load()

◆ cond_neural_match()

◆ cond_neural_mutate()

◆ cond_neural_neurons()

◆ cond_neural_param_defaults()

◆ cond_neural_param_json_import()

◆ cond_neural_print()

◆ cond_neural_save()

◆ cond_neural_size()

◆ cond_neural_update()