Defines the interface for a learning algorithm. More...

#include <reductions_fwd.h>

Public Types
using	end_fptr_type = void()(VW::workspace &, void , void *)

using	finish_fptr_type = void()(void )

Public Member Functions
void *	get_internal_type_erased_data_pointer_test_use_only ()

void	learn (E &ec, size_t i=0)
	Will update the model according to the labels and examples supplied. More...

void	predict (E &ec, size_t i=0)
	Make a prediction for the given example. More...

void	multipredict (E &ec, size_t lo, size_t count, polyprediction *pred, bool finalize_predictions)

void	update (E &ec, size_t i=0)

float	sensitivity (example &ec, size_t i=0)

void	save_load (io_buf &io, const bool read, const bool text)

void	persist_metrics (metric_sink &metrics)

void	finish ()

void	end_pass ()

void	end_examples ()

void	init_driver ()

void	finish_example (VW::workspace &all, E &ec)

void	print_example (VW::workspace &all, E &ec)

void	get_enabled_reductions (std::vector< std::string > &enabled_reductions) const

base_learner *	get_learner_by_name_prefix (const std::string &reduction_name)

prediction_type_t	get_output_prediction_type ()

prediction_type_t	get_input_prediction_type ()

label_type_t	get_output_label_type ()

label_type_t	get_input_label_type ()

bool	is_multiline ()

const std::string &	get_name ()

Public Attributes
size_t	weights

size_t	increment

bool	learn_returns_prediction = false

Friends
template<class FluentBuilderT , class DataT , class ExampleT , class BaseLearnerT >
struct	common_learner_builder

template<class DataT , class ExampleT >
struct	base_learner_builder

template<class DataT , class ExampleT , class BaseLearnerT >
struct	reduction_learner_builder

template<class ExampleT , class BaseLearnerT >
struct	reduction_no_data_learner_builder

Detailed Description

template<class T, class E>
struct VW::LEARNER::learner< T, E >

Defines the interface for a learning algorithm.

Learner is implemented as a struct of pointers, and associated methods. It implements a sort of virtual inheritance through the use of bundling function pointers with the associated objects to call them with. A reduction will recursively call the base given to it, whereas a base learner will not recurse and will simply return the result. Learner is not intended to be inherited from. Instead it is used through composition, and created through the make_reduction_learner and make_base_learner. The state of this learner, or reduction, is stored in the learner_data field. A std::shared_pointer<void> is used as this class uses type erasure to allow for an arbitrary reduction to be implemented. It is extremely important that the function pointers given to the class match the expected types of the object. If the learner is constructed using make_reduction_learner or make_base_learner and assembled before it is transformed into a VW::LEARNER::base_learner with VW::LEARNER::make_base then the usage of the templated functions should ensure types are correct.

Template Parameters

T	Type of the reduction data object stored. This allows this specific reduction to have it's own state.
E	Example type this reduction supports. Must be one of example or multi_ex

Member Typedef Documentation

◆ end_fptr_type

template<class T, class E>

using VW::LEARNER::learner< T, E >::end_fptr_type = void (*)(VW::workspace&, void*, void*)

◆ finish_fptr_type

template<class T, class E>

using VW::LEARNER::learner< T, E >::finish_fptr_type = void (*)(void*)

Member Function Documentation

◆ end_examples()

template<class T, class E>

void VW::LEARNER::learner< T, E >::end_examples ( )

inline

◆ end_pass()

template<class T, class E>

void VW::LEARNER::learner< T, E >::end_pass ( )

inline

◆ finish()

template<class T, class E>

void VW::LEARNER::learner< T, E >::finish ( )

inline

◆ finish_example()

template<class T, class E>

void VW::LEARNER::learner< T, E >::finish_example	(	VW::workspace &	all,
		E &	ec
	)

inline

◆ get_enabled_reductions()

template<class T, class E>

void VW::LEARNER::learner< T, E >::get_enabled_reductions ( std::vector< std::string > & enabled_reductions ) const

inline

◆ get_input_label_type()

template<class T, class E>

label_type_t VW::LEARNER::learner< T, E >::get_input_label_type ( )

inline

◆ get_input_prediction_type()

template<class T, class E>

prediction_type_t VW::LEARNER::learner< T, E >::get_input_prediction_type ( )

inline

◆ get_internal_type_erased_data_pointer_test_use_only()

template<class T, class E>

void* VW::LEARNER::learner< T, E >::get_internal_type_erased_data_pointer_test_use_only ( )

inline

◆ get_learner_by_name_prefix()

template<class T, class E>

base_learner* VW::LEARNER::learner< T, E >::get_learner_by_name_prefix ( const std::string & reduction_name )

inline

◆ get_name()

template<class T, class E>

const std::string& VW::LEARNER::learner< T, E >::get_name ( )

inline

◆ get_output_label_type()

template<class T, class E>

label_type_t VW::LEARNER::learner< T, E >::get_output_label_type ( )

inline

◆ get_output_prediction_type()

template<class T, class E>

prediction_type_t VW::LEARNER::learner< T, E >::get_output_prediction_type ( )

inline

◆ init_driver()

template<class T, class E>

void VW::LEARNER::learner< T, E >::init_driver ( )

inline

◆ is_multiline()

template<class T, class E>

bool VW::LEARNER::learner< T, E >::is_multiline ( )

inline

◆ learn()

template<class T, class E>

void VW::LEARNER::learner< T, E >::learn	(	E &	ec,
		size_t	i = `0`
	)

inline

Will update the model according to the labels and examples supplied.

Parameters

ec	The example object or multi_ex to be operated on. This object must have a valid label set for every example in the field example::l that corresponds to the type this reduction expects.
i	This is the offset used for the weights in this call. If using multiple regressors/learners you can increment this value for each call.

Returns: While some reductions may fill the example::pred, this is not guaranteed and is undefined behavior if accessed.

◆ multipredict()

template<class T, class E>

void VW::LEARNER::learner< T, E >::multipredict	(	E &	ec,
		size_t	lo,
		size_t	count,
		polyprediction *	pred,
		bool	finalize_predictions
	)

inline

◆ persist_metrics()

template<class T, class E>

void VW::LEARNER::learner< T, E >::persist_metrics ( metric_sink & metrics )

inline

◆ predict()

template<class T, class E>

void VW::LEARNER::learner< T, E >::predict	(	E &	ec,
		size_t	i = `0`
	)

inline

Make a prediction for the given example.

Parameters

ec	The example object or multi_ex to be operated on. This object must have a valid prediction allocated in the field example::pred that corresponds to this reduction type.
i	This is the offset used for the weights in this call. If using multiple regressors/learners you can increment this value for each call.

Returns: The prediction calculated by this reduction be set on example::pred. If E is multi_ex then the prediction is set on the 0th item in the list.

◆ print_example()

template<class T, class E>

void VW::LEARNER::learner< T, E >::print_example	(	VW::workspace &	all,
		E &	ec
	)

inline

◆ save_load()

template<class T, class E>

void VW::LEARNER::learner< T, E >::save_load	(	io_buf &	io,
		const bool	read,
		const bool	text
	)

inline

◆ sensitivity()

template<class T, class E>

float VW::LEARNER::learner< T, E >::sensitivity	(	example &	ec,
		size_t	i = `0`
	)

inline

◆ update()

template<class T, class E>

void VW::LEARNER::learner< T, E >::update	(	E &	ec,
		size_t	i = `0`
	)

inline

Friends And Related Function Documentation

◆ base_learner_builder

template<class T, class E>

template<class DataT , class ExampleT >

friend struct base_learner_builder

friend

◆ common_learner_builder

template<class T, class E>

template<class FluentBuilderT , class DataT , class ExampleT , class BaseLearnerT >

friend struct common_learner_builder

friend

◆ reduction_learner_builder

template<class T, class E>

template<class DataT , class ExampleT , class BaseLearnerT >

friend struct reduction_learner_builder

friend

◆ reduction_no_data_learner_builder

template<class T, class E>

template<class ExampleT , class BaseLearnerT >

friend struct reduction_no_data_learner_builder

friend

Member Data Documentation

◆ increment

template<class T, class E>

size_t VW::LEARNER::learner< T, E >::increment

◆ learn_returns_prediction

template<class T, class E>

bool VW::LEARNER::learner< T, E >::learn_returns_prediction = false

◆ weights

template<class T, class E>

size_t VW::LEARNER::learner< T, E >::weights

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

/__w/vowpal_wabbit/vowpal_wabbit/vowpalwabbit/learner.h

Public Types

Public Member Functions

Public Attributes

Friends

Detailed Description

template<class T, class E> struct VW::LEARNER::learner< T, E >

Member Typedef Documentation

◆ end_fptr_type

◆ finish_fptr_type

Member Function Documentation

◆ end_examples()

◆ end_pass()

◆ finish()

◆ finish_example()

◆ get_enabled_reductions()

◆ get_input_label_type()

◆ get_input_prediction_type()

◆ get_internal_type_erased_data_pointer_test_use_only()

◆ get_learner_by_name_prefix()

◆ get_name()

◆ get_output_label_type()

◆ get_output_prediction_type()

◆ init_driver()

◆ is_multiline()

◆ learn()

◆ multipredict()

◆ persist_metrics()

◆ predict()

◆ print_example()

◆ save_load()

◆ sensitivity()

◆ update()

Friends And Related Function Documentation

◆ base_learner_builder

◆ common_learner_builder

◆ reduction_learner_builder

◆ reduction_no_data_learner_builder

Member Data Documentation

◆ increment

◆ learn_returns_prediction

◆ weights

template<class T, class E>
struct VW::LEARNER::learner< T, E >