Python XCSF class data structure. More...

Collaboration diagram for XCS:

Public Member Functions
	XCS ()
	Default Constructor. More...

	XCS (py::kwargs kwargs)
	Constructor. More...

void	reset (void)
	Resets basic constructor variables. More...

size_t	save (const char *filename)
	Writes the entire current state of XCSF to a file. More...

size_t	load (const char *filename)
	Reads the entire current state of XCSF from a file. More...

void	store (void)
	Stores the current population in memory for later retrieval. More...

void	retrieve (void)
	Retrieves the stored population, setting it as current. More...

void	print_params (void)
	Prints the XCSF parameters and their current values. More...

void	pred_expand (void)
	Inserts a new hidden layer before the output layer within all prediction neural networks in the population. More...

void	ae_to_classifier (const int y_dim, const int n_del)
	Switches from autoencoding to classification. More...

void	print_pset (const bool condition, const bool action, const bool prediction)
	Prints the current population. More...

double	fit (const py::array_t< double > input, const int action, const double reward)
	Creates/updates an action set for a given (state, action, reward). More...

void	init_trial (void)
	Initialises a reinforcement learning trial. More...

void	end_trial (void)
	Frees memory used by a reinforcement learning trial. More...

void	init_step (void)
	Initialises a step in a reinforcement learning trial. More...

void	end_step (void)
	Ends a step in a reinforcement learning trial. More...

int	decision (const py::array_t< double > input, const bool explore)
	Selects an action to perform in a reinforcement learning problem. More...

void	update (const double reward, const bool done)
	Creates the action set using the previously selected action, updates the classifiers, and runs the EA on explore steps. More...

double	error (const double reward, const bool done, const double max_p)
	Returns the reinforcement learning system prediction error. More...

void	load_input (struct Input *data, const py::array_t< double > X, const py::array_t< double > Y)
	Loads an input data structure for fitting. More...

void	print_status ()
	Prints the current performance metrics. More...

void	update_metrics (const double train, const double val, const int n_trials)
	Updates performance metrics. More...

void	load_validation_data (py::kwargs kwargs)
	Loads validation data if present in kwargs. More...

bool	callbacks_run (py::list callbacks)
	Executes callbacks and returns whether to terminate. More...

void	callbacks_finish (py::list callbacks)
	Executes callback finish. More...

XCS &	fit (const py::array_t< double > X_train, const py::array_t< double > y_train, const bool shuffle, const bool warm_start, const bool verbose, py::object callbacks, py::kwargs kwargs)
	Executes MAX_TRIALS number of XCSF learning iterations using the provided training data. More...

double *	get_cover (const py::array_t< double > cover)
	Returns the values specified in the cover array. More...

void	set_cover (const py::object &cover)
	Sets the XCSF cover array to values given, or zeros. More...

py::array_t< double >	predict (const py::array_t< double > X, const py::object &cover)
	Returns the XCSF prediction array for the provided input. More...

double	score (const py::array_t< double > X, const py::array_t< double > Y, const int N, const py::object &cover)
	Returns the error using N random samples from the provided data. More...

py::bytes	serialize () const
	Implements pickle file writing. More...

double	error (void)
	Returns the current system error. More...

py::dict	get_metrics (void)

int	get_pset_size (void)

int	get_pset_num (void)

int	get_time (void)

double	get_pset_mean_cond_size (void)

double	get_pset_mean_pred_size (void)

double	get_pset_mean_pred_eta (const int layer)

double	get_pset_mean_pred_neurons (const int layer)

double	get_pset_mean_pred_connections (const int layer)

double	get_pset_mean_pred_layers (void)

double	get_pset_mean_cond_connections (const int layer)

double	get_pset_mean_cond_neurons (const int layer)

double	get_pset_mean_cond_layers (void)

double	get_mset_size (void)

double	get_aset_size (void)

double	get_mfrac (void)

const char *	json_export (const bool condition, const bool action, const bool prediction)
	Returns a JSON formatted string representing the population set. More...

void	update_params ()
	Updates the Python object's parameter dictionary. More...

py::dict	get_params (const bool deep)
	Returns a dictionary of parameters. More...

XCS &	set_params (py::kwargs kwargs)
	Sets parameter values. More...

py::dict	internal_params ()
	Returns a dictionary of the internal parameters. More...

void	json_insert_cl (const std::string &json_str)
	Creates a classifier from JSON and inserts into the population. More...

void	json_insert (const std::string &json_str)
	Creates classifiers from JSON and inserts into the population. More...

void	json_write (const std::string &filename)
	Writes the current population set to a file in JSON. More...

void	json_read (const std::string &filename)
	Reads classifiers from a JSON file and adds to the population. More...

Static Public Member Functions
static XCS	deserialize (const py::bytes &state)
	Implements pickle file reading. More...

Private Attributes
struct XCSF	xcs
	XCSF data structure. More...

double *	state
	Current input state for RL. More...

int	action
	Current action for RL. More...

double	payoff
	Current reward for RL. More...

struct Input *	train_data
	Training data for supervised learning. More...

struct Input *	test_data
	Test data for supervised learning. More...

struct Input *	val_data
	Validation data. More...

py::dict	params
	Dictionary of parameters and their values. More...

py::list	metric_train

py::list	metric_val

py::list	metric_trial

py::list	metric_psize

py::list	metric_msize

py::list	metric_mfrac

int	metric_counter

Detailed Description

Python XCSF class data structure.

Definition at line 64 of file pybind_wrapper.cpp.

Constructor & Destructor Documentation

◆ XCS() [1/2]

XCS::XCS ( )

inline

Default Constructor.

Definition at line 87 of file pybind_wrapper.cpp.

References reset(), xcs, and xcsf_init().

Referenced by deserialize().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ XCS() [2/2]

XCS::XCS ( py::kwargs kwargs )

inlineexplicit

Constructor.

Parameters

[in] kwargs Parameters and their values.

Definition at line 97 of file pybind_wrapper.cpp.

References reset(), set_params(), xcs, and xcsf_init().

Here is the call graph for this function:

Member Function Documentation

◆ ae_to_classifier()

void XCS::ae_to_classifier	(	const int	y_dim,
		const int	n_del
	)

inline

Switches from autoencoding to classification.

Parameters

[in]	y_dim	The output dimension (i.e., the number of classes).
[in]	n_del	The number of hidden layers to remove.

Definition at line 198 of file pybind_wrapper.cpp.

References xcs, xcsf_ae_to_classifier(), and Input::y_dim.

Here is the call graph for this function:

◆ callbacks_finish()

void XCS::callbacks_finish ( py::list callbacks )

inline

Executes callback finish.

Parameters

[in] callbacks The callbacks to perform.

Definition at line 474 of file pybind_wrapper.cpp.

References Callback::finish(), and xcs.

Referenced by fit().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ callbacks_run()

bool XCS::callbacks_run ( py::list callbacks )

inline

Executes callbacks and returns whether to terminate.

Parameters

[in] callbacks The callbacks to perform.

Returns: Whether to terminate early.

Definition at line 452 of file pybind_wrapper.cpp.

References get_metrics(), Callback::run(), and xcs.

Referenced by fit().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ decision()

int XCS::decision	(	const py::array_t< double >	input,
		const bool	explore
	)

inline

Selects an action to perform in a reinforcement learning problem.

Constructs the match set and selects an action to perform.

Parameters

[in]	input	The input state.
[in]	explore	Whether this is an exploration step.

Returns: The selected action.

Definition at line 287 of file pybind_wrapper.cpp.

References action, error(), param_set_explore(), state, XCSF::x_dim, xcs, and xcs_rl_decision().

Here is the call graph for this function:

◆ deserialize()

static XCS XCS::deserialize ( const py::bytes & state )

inlinestatic

Implements pickle file reading.

Uses a temporary binary file.

Parameters

state The pickled state of a saved XCSF.

Definition at line 659 of file pybind_wrapper.cpp.

References state, XCS(), xcs, and xcsf_load().

Here is the call graph for this function:

◆ end_step()

void XCS::end_step ( void )

inline

Ends a step in a reinforcement learning trial.

Definition at line 274 of file pybind_wrapper.cpp.

References action, payoff, state, xcs, and xcs_rl_end_step().

Here is the call graph for this function:

◆ end_trial()

void XCS::end_trial ( void )

inline

Frees memory used by a reinforcement learning trial.

Definition at line 256 of file pybind_wrapper.cpp.

References xcs, and xcs_rl_end_trial().

Here is the call graph for this function:

◆ error() [1/2]

double XCS::error	(	const double	reward,
		const bool	done,
		const double	max_p
	)

inline

Returns the reinforcement learning system prediction error.

Parameters

[in]	reward	The current reward.
[in]	done	Whether the environment is in a terminal state.
[in]	max_p	The maximum payoff in the environment.

Returns: The prediction error.

Definition at line 322 of file pybind_wrapper.cpp.

References action, payoff, xcs, and xcs_rl_error().

Here is the call graph for this function:

◆ error() [2/2]

double XCS::error ( void )

inline

Returns the current system error.

Returns: Moving average of the system error, updated with step size BETA.

Definition at line 688 of file pybind_wrapper.cpp.

References XCSF::error, and xcs.

Referenced by decision(), fit(), load_input(), and predict().

Here is the caller graph for this function:

◆ fit() [1/2]

double XCS::fit	(	const py::array_t< double >	input,
		const int	action,
		const double	reward
	)

inline

Creates/updates an action set for a given (state, action, reward).

Parameters

[in]	input	The input state to match.
[in]	action	The selected action.
[in]	reward	The reward for having performed the action.

Returns: The prediction error.

Definition at line 225 of file pybind_wrapper.cpp.

References action, error(), XCSF::n_actions, state, XCSF::x_dim, xcs, and xcs_rl_fit().

Here is the call graph for this function:

◆ fit() [2/2]

XCS& XCS::fit	(	const py::array_t< double >	X_train,
		const py::array_t< double >	y_train,
		const bool	shuffle,
		const bool	warm_start,
		const bool	verbose,
		py::object	callbacks,
		py::kwargs	kwargs
	)

inline

Executes MAX_TRIALS number of XCSF learning iterations using the provided training data.

Parameters

[in]	X_train	The input values to use for training.
[in]	y_train	The true output values to use for training.
[in]	shuffle	Whether to randomise the instances during training.
[in]	warm_start	Whether to continue with existing population.
[in]	verbose	Whether to print learning metrics.
[in]	callbacks	List of Callback objects or None.
[in]	kwargs	Keyword arguments.

Returns: The fitted XCSF model.

Definition at line 499 of file pybind_wrapper.cpp.

References callbacks_finish(), callbacks_run(), XCSF::cover, load_input(), load_validation_data(), XCSF::MAX_TRIALS, XCSF::PERF_TRIALS, print_status(), train_data, update_metrics(), val_data, xcs, xcs_supervised_fit(), xcs_supervised_score(), xcsf_free(), and xcsf_init().

Here is the call graph for this function:

◆ get_aset_size()

double XCS::get_aset_size ( void )

inline

Definition at line 785 of file pybind_wrapper.cpp.

References XCSF::aset_size, and xcs.

◆ get_cover()

double* XCS::get_cover ( const py::array_t< double > cover )

inline

Returns the values specified in the cover array.

Parameters

[in] cover The values to return for covering.

Returns: The cover array values.

Definition at line 543 of file pybind_wrapper.cpp.

References xcs, and XCSF::y_dim.

Referenced by set_cover().

Here is the caller graph for this function:

◆ get_metrics()

py::dict XCS::get_metrics ( void )

inline

Definition at line 694 of file pybind_wrapper.cpp.

References metric_mfrac, metric_msize, metric_psize, metric_train, metric_trial, and metric_val.

Referenced by callbacks_run().

Here is the caller graph for this function:

◆ get_mfrac()

double XCS::get_mfrac ( void )

inline

Definition at line 791 of file pybind_wrapper.cpp.

References XCSF::mfrac, and xcs.

◆ get_mset_size()

double XCS::get_mset_size ( void )

inline

Definition at line 779 of file pybind_wrapper.cpp.

References XCSF::mset_size, and xcs.

◆ get_params()

py::dict XCS::get_params ( const bool deep )

inline

Returns a dictionary of parameters.

Parameters

deep	For sklearn compatibility.

Returns: External parameter dictionary.

Definition at line 842 of file pybind_wrapper.cpp.

References params.

◆ get_pset_mean_cond_connections()

double XCS::get_pset_mean_cond_connections ( const int layer )

inline

Definition at line 761 of file pybind_wrapper.cpp.

References clset_mean_cond_connections(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_cond_layers()

double XCS::get_pset_mean_cond_layers ( void )

inline

Definition at line 773 of file pybind_wrapper.cpp.

References clset_mean_cond_layers(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_cond_neurons()

double XCS::get_pset_mean_cond_neurons ( const int layer )

inline

Definition at line 767 of file pybind_wrapper.cpp.

References clset_mean_cond_neurons(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_cond_size()

double XCS::get_pset_mean_cond_size ( void )

inline

Definition at line 725 of file pybind_wrapper.cpp.

References clset_mean_cond_size(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_pred_connections()

double XCS::get_pset_mean_pred_connections ( const int layer )

inline

Definition at line 749 of file pybind_wrapper.cpp.

References clset_mean_pred_connections(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_pred_eta()

double XCS::get_pset_mean_pred_eta ( const int layer )

inline

Definition at line 737 of file pybind_wrapper.cpp.

References clset_mean_pred_eta(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_pred_layers()

double XCS::get_pset_mean_pred_layers ( void )

inline

Definition at line 755 of file pybind_wrapper.cpp.

References clset_mean_pred_layers(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_pred_neurons()

double XCS::get_pset_mean_pred_neurons ( const int layer )

inline

Definition at line 743 of file pybind_wrapper.cpp.

References clset_mean_pred_neurons(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_mean_pred_size()

double XCS::get_pset_mean_pred_size ( void )

inline

Definition at line 731 of file pybind_wrapper.cpp.

References clset_mean_pred_size(), XCSF::pset, and xcs.

Here is the call graph for this function:

◆ get_pset_num()

int XCS::get_pset_num ( void )

inline

Definition at line 713 of file pybind_wrapper.cpp.

References Set::num, XCSF::pset, and xcs.

◆ get_pset_size()

int XCS::get_pset_size ( void )

inline

Definition at line 707 of file pybind_wrapper.cpp.

References XCSF::pset, Set::size, and xcs.

◆ get_time()

int XCS::get_time ( void )

inline

Definition at line 719 of file pybind_wrapper.cpp.

References XCSF::time, and xcs.

◆ init_step()

void XCS::init_step ( void )

inline

Initialises a step in a reinforcement learning trial.

Definition at line 265 of file pybind_wrapper.cpp.

References xcs, and xcs_rl_init_step().

Here is the call graph for this function:

◆ init_trial()

void XCS::init_trial ( void )

inline

Initialises a reinforcement learning trial.

Definition at line 247 of file pybind_wrapper.cpp.

References xcs, and xcs_rl_init_trial().

Here is the call graph for this function:

◆ internal_params()

py::dict XCS::internal_params ( )

inline

Returns a dictionary of the internal parameters.

Returns: Internal parameter dictionary.

Definition at line 882 of file pybind_wrapper.cpp.

References param_json_export(), and xcs.

Here is the call graph for this function:

◆ json_export()

const char* XCS::json_export	(	const bool	condition,
		const bool	action,
		const bool	prediction
	)

inline

Returns a JSON formatted string representing the population set.

Parameters

[in]	condition	Whether to return the condition.
[in]	action	Whether to return the action.
[in]	prediction	Whether to return the prediction.

Returns: String encoded in json format.

Definition at line 806 of file pybind_wrapper.cpp.

References action, clset_json_export(), Set::list, XCSF::pset, and xcs.

Referenced by json_write().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ json_insert()

void XCS::json_insert ( const std::string & json_str )

inline

Creates classifiers from JSON and inserts into the population.

Parameters

[in] json_str JSON formatted string representing a classifier.

Definition at line 909 of file pybind_wrapper.cpp.

References clset_json_insert(), and xcs.

Referenced by json_read().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ json_insert_cl()

void XCS::json_insert_cl ( const std::string & json_str )

inline

Creates a classifier from JSON and inserts into the population.

Parameters

[in] json_str JSON formatted string representing a classifier.

Definition at line 896 of file pybind_wrapper.cpp.

References clset_json_insert_cl(), utils_json_parse_check(), and xcs.

Here is the call graph for this function:

◆ json_read()

void XCS::json_read ( const std::string & filename )

inline

Reads classifiers from a JSON file and adds to the population.

Parameters

[in] filename Name of the input file.

Definition at line 931 of file pybind_wrapper.cpp.

References json_insert().

Here is the call graph for this function:

◆ json_write()

void XCS::json_write ( const std::string & filename )

inline

Writes the current population set to a file in JSON.

Parameters

[in] filename Name of the output file.

Definition at line 919 of file pybind_wrapper.cpp.

References json_export().

Here is the call graph for this function:

◆ load()

size_t XCS::load ( const char * filename )

inline

Reads the entire current state of XCSF from a file.

Parameters

[in] filename String containing the name of the input file.

Returns: The total number of elements read.

Definition at line 148 of file pybind_wrapper.cpp.

References update_params(), xcs, and xcsf_load().

Here is the call graph for this function:

◆ load_input()

void XCS::load_input	(	struct Input *	data,
		const py::array_t< double >	X,
		const py::array_t< double >	Y
	)

inline

Loads an input data structure for fitting.

Parameters

[in,out]	data	Input data structure used to point to the data.
[in]	X	Vector of features with shape (n_samples, x_dim).
[in]	Y	Vector of truth values with shape (n_samples, y_dim).

Definition at line 337 of file pybind_wrapper.cpp.

References error(), Input::n_samples, Input::x, XCSF::x_dim, Input::x_dim, xcs, Input::y, XCSF::y_dim, and Input::y_dim.

Referenced by fit(), load_validation_data(), and score().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ load_validation_data()

void XCS::load_validation_data ( py::kwargs kwargs )

inline

Loads validation data if present in kwargs.

Parameters

[in] kwargs Parameters and their values.

Definition at line 426 of file pybind_wrapper.cpp.

References XCSF::cover, load_input(), XCSF::pa_size, test_data, val_data, and xcs.

Referenced by fit().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ pred_expand()

void XCS::pred_expand ( void )

inline

Inserts a new hidden layer before the output layer within all prediction neural networks in the population.

Definition at line 187 of file pybind_wrapper.cpp.

References xcs, and xcsf_pred_expand().

Here is the call graph for this function:

◆ predict()

py::array_t<double> XCS::predict	(	const py::array_t< double >	X,
		const py::object &	cover
	)

inline

Returns the XCSF prediction array for the provided input.

Parameters

[in]	X	The input variables.
[in]	cover	If the match set is empty, the prediction array will be set to this value instead of covering.

Returns: The prediction array values.

Definition at line 584 of file pybind_wrapper.cpp.

References XCSF::cover, error(), Input::n_samples, XCSF::pa_size, set_cover(), XCSF::x_dim, xcs, and xcs_supervised_predict().

Here is the call graph for this function:

◆ print_params()

void XCS::print_params ( void )

inline

Prints the XCSF parameters and their current values.

Definition at line 177 of file pybind_wrapper.cpp.

References param_print(), and xcs.

Here is the call graph for this function:

◆ print_pset()

void XCS::print_pset	(	const bool	condition,
		const bool	action,
		const bool	prediction
	)

inline

Prints the current population.

Parameters

[in]	condition	Whether to print the condition.
[in]	action	Whether to print the action.
[in]	prediction	Whether to print the prediction.

Definition at line 210 of file pybind_wrapper.cpp.

References action, xcs, and xcsf_print_pset().

Here is the call graph for this function:

◆ print_status()

void XCS::print_status ( )

inline

Prints the current performance metrics.

Definition at line 381 of file pybind_wrapper.cpp.

References get_timestamp(), metric_mfrac, metric_msize, metric_psize, metric_train, metric_trial, metric_val, and val_data.

Referenced by fit().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ reset()

void XCS::reset ( void )

inline

Resets basic constructor variables.

Definition at line 108 of file pybind_wrapper.cpp.

References action, metric_counter, Input::n_samples, param_init(), payoff, state, test_data, train_data, update_params(), val_data, Input::x, Input::x_dim, xcs, Input::y, and Input::y_dim.

Referenced by XCS().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ retrieve()

void XCS::retrieve ( void )

inline

Retrieves the stored population, setting it as current.

Definition at line 168 of file pybind_wrapper.cpp.

References xcs, and xcsf_retrieve_pset().

Here is the call graph for this function:

◆ save()

size_t XCS::save ( const char * filename )

inline

Writes the entire current state of XCSF to a file.

Parameters

[in] filename String containing the name of the output file.

Returns: The total number of elements written.

Definition at line 137 of file pybind_wrapper.cpp.

References xcs, and xcsf_save().

Here is the call graph for this function:

◆ score()

double XCS::score	(	const py::array_t< double >	X,
		const py::array_t< double >	Y,
		const int	N,
		const py::object &	cover
	)

inline

Returns the error using N random samples from the provided data.

Parameters

[in]	X	The input values to use for scoring.
[in]	Y	The true output values to use for scoring.
[in]	N	The maximum number of samples to draw randomly for scoring.
[in]	cover	If the match set is empty, the prediction array will be set to this value, otherwise it is set to zeros.

Returns: The average XCSF error using the loss function.

Definition at line 618 of file pybind_wrapper.cpp.

References XCSF::cover, load_input(), set_cover(), test_data, xcs, xcs_supervised_score(), and xcs_supervised_score_n().

Here is the call graph for this function:

◆ serialize()

py::bytes XCS::serialize ( ) const

inline

Implements pickle file writing.

Uses a temporary binary file.

Returns: The pickled XCSF.

Definition at line 635 of file pybind_wrapper.cpp.

References state, xcs, and xcsf_save().

Here is the call graph for this function:

◆ set_cover()

void XCS::set_cover ( const py::object & cover )

inline

Sets the XCSF cover array to values given, or zeros.

Parameters

[in] cover The values to use instead of covering.

Definition at line 566 of file pybind_wrapper.cpp.

References XCSF::cover, get_cover(), XCSF::pa_size, and xcs.

Referenced by predict(), and score().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ set_params()

XCS& XCS::set_params ( py::kwargs kwargs )

inline

Sets parameter values.

Parameters

kwargs Parameters and their values.

Returns: The XCSF object.

Definition at line 854 of file pybind_wrapper.cpp.

References param_json_import(), params, and xcs.

Referenced by XCS().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ store()

void XCS::store ( void )

inline

Stores the current population in memory for later retrieval.

Definition at line 159 of file pybind_wrapper.cpp.

References xcs, and xcsf_store_pset().

Here is the call graph for this function:

◆ update()

void XCS::update	(	const double	reward,
		const bool	done
	)

inline

Creates the action set using the previously selected action, updates the classifiers, and runs the EA on explore steps.

Parameters

[in]	reward	The reward from performing the action.
[in]	done	Whether the environment is in a terminal state.

Definition at line 308 of file pybind_wrapper.cpp.

References action, payoff, state, xcs, and xcs_rl_update().

Here is the call graph for this function:

◆ update_metrics()

void XCS::update_metrics	(	const double	train,
		const double	val,
		const int	n_trials
	)

inline

Updates performance metrics.

Parameters

[in]	train	The current training error.
[in]	val	The current validation error.
[in]	n_trials	Number of trials run.

Definition at line 409 of file pybind_wrapper.cpp.

References metric_counter, metric_mfrac, metric_msize, metric_psize, metric_train, metric_trial, metric_val, XCSF::mfrac, XCSF::mset_size, XCSF::pset, Set::size, and xcs.

Referenced by fit().

Here is the caller graph for this function:

◆ update_params()

void XCS::update_params ( )

inline

Updates the Python object's parameter dictionary.

Definition at line 819 of file pybind_wrapper.cpp.

References param_json_export(), params, and xcs.

Referenced by load(), and reset().

Here is the call graph for this function:

Here is the caller graph for this function:

Field Documentation

◆ action

int XCS::action

private

Current action for RL.

Definition at line 69 of file pybind_wrapper.cpp.

Referenced by decision(), end_step(), error(), fit(), json_export(), print_pset(), reset(), and update().

◆ metric_counter

int XCS::metric_counter

private

Definition at line 81 of file pybind_wrapper.cpp.

Referenced by reset(), and update_metrics().

◆ metric_mfrac

py::list XCS::metric_mfrac

private

Definition at line 80 of file pybind_wrapper.cpp.

Referenced by get_metrics(), print_status(), and update_metrics().

◆ metric_msize

py::list XCS::metric_msize

private

Definition at line 79 of file pybind_wrapper.cpp.

Referenced by get_metrics(), print_status(), and update_metrics().

◆ metric_psize

py::list XCS::metric_psize

private

Definition at line 78 of file pybind_wrapper.cpp.

Referenced by get_metrics(), print_status(), and update_metrics().

◆ metric_train

py::list XCS::metric_train

private

Definition at line 75 of file pybind_wrapper.cpp.

Referenced by get_metrics(), print_status(), and update_metrics().

◆ metric_trial

py::list XCS::metric_trial

private

Definition at line 77 of file pybind_wrapper.cpp.

Referenced by get_metrics(), print_status(), and update_metrics().

◆ metric_val

py::list XCS::metric_val

private

Definition at line 76 of file pybind_wrapper.cpp.

Referenced by get_metrics(), print_status(), and update_metrics().

◆ params

py::dict XCS::params

private

Dictionary of parameters and their values.

Definition at line 74 of file pybind_wrapper.cpp.

Referenced by get_params(), set_params(), and update_params().

◆ payoff

double XCS::payoff

private

Current reward for RL.

Definition at line 70 of file pybind_wrapper.cpp.

Referenced by end_step(), error(), reset(), and update().

◆ state

double* XCS::state

private

Current input state for RL.

Definition at line 68 of file pybind_wrapper.cpp.

Referenced by decision(), deserialize(), end_step(), fit(), reset(), serialize(), and update().

◆ test_data

struct Input* XCS::test_data

private

Test data for supervised learning.

Definition at line 72 of file pybind_wrapper.cpp.

Referenced by load_validation_data(), reset(), and score().

◆ train_data

struct Input* XCS::train_data

private

Training data for supervised learning.

Definition at line 71 of file pybind_wrapper.cpp.

Referenced by fit(), and reset().

◆ val_data

struct Input* XCS::val_data

private

Validation data.

Definition at line 73 of file pybind_wrapper.cpp.

Referenced by fit(), load_validation_data(), print_status(), and reset().

◆ xcs

struct XCSF XCS::xcs

private

XCSF data structure.

Definition at line 940 of file pybind_wrapper.cpp.

Referenced by ae_to_classifier(), callbacks_finish(), callbacks_run(), decision(), deserialize(), end_step(), end_trial(), error(), fit(), get_aset_size(), get_cover(), get_mfrac(), get_mset_size(), get_pset_mean_cond_connections(), get_pset_mean_cond_layers(), get_pset_mean_cond_neurons(), get_pset_mean_cond_size(), get_pset_mean_pred_connections(), get_pset_mean_pred_eta(), get_pset_mean_pred_layers(), get_pset_mean_pred_neurons(), get_pset_mean_pred_size(), get_pset_num(), get_pset_size(), get_time(), init_step(), init_trial(), internal_params(), json_export(), json_insert(), json_insert_cl(), load(), load_input(), load_validation_data(), pred_expand(), predict(), print_params(), print_pset(), reset(), retrieve(), save(), score(), serialize(), set_cover(), set_params(), store(), update(), update_metrics(), update_params(), and XCS().

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

pybind_wrapper.cpp

Public Member Functions

Static Public Member Functions

Private Attributes

Detailed Description

Constructor & Destructor Documentation

◆ XCS() [1/2]

◆ XCS() [2/2]

Member Function Documentation

◆ ae_to_classifier()

◆ callbacks_finish()

◆ callbacks_run()

◆ decision()

◆ deserialize()

◆ end_step()

◆ end_trial()

◆ error() [1/2]

◆ error() [2/2]

◆ fit() [1/2]

◆ fit() [2/2]

◆ get_aset_size()

◆ get_cover()

◆ get_metrics()

◆ get_mfrac()

◆ get_mset_size()

◆ get_params()

◆ get_pset_mean_cond_connections()

◆ get_pset_mean_cond_layers()

◆ get_pset_mean_cond_neurons()

◆ get_pset_mean_cond_size()

◆ get_pset_mean_pred_connections()

◆ get_pset_mean_pred_eta()

◆ get_pset_mean_pred_layers()

◆ get_pset_mean_pred_neurons()

◆ get_pset_mean_pred_size()

◆ get_pset_num()

◆ get_pset_size()

◆ get_time()

◆ init_step()

◆ init_trial()

◆ internal_params()

◆ json_export()

◆ json_insert()

◆ json_insert_cl()

◆ json_read()

◆ json_write()

◆ load()

◆ load_input()

◆ load_validation_data()

◆ pred_expand()

◆ predict()

◆ print_params()

◆ print_pset()

◆ print_status()

◆ reset()

◆ retrieve()

◆ save()

◆ score()

◆ serialize()

◆ set_cover()

◆ set_params()

◆ store()

◆ update()

◆ update_metrics()

◆ update_params()

Field Documentation

◆ action

◆ metric_counter

◆ metric_mfrac

◆ metric_msize

◆ metric_psize

◆ metric_train

◆ metric_trial

◆ metric_val

◆ params

◆ payoff

◆ state

◆ test_data

◆ train_data

◆ val_data

◆ xcs