class Trainer

Trainer.

Inheritance:

Public Methods

Trainer(Machine* machine_)

virtual void train(DataSet* data_, MeasurerList* measurers) = 0

Train the machine.

virtual void test(MeasurerList* measurers)

Test the machine.

static Allocator* extractMeasurers(MeasurerList* measurers, DataSet* train, DataSet*** datas, Measurer**** meas, int** n_meas, int* n_datas)

Make a table of measurers from a List.

virtual void loadXFile(XFile* file)

By default, just load the machine

virtual void saveXFile(XFile* file)

By default, just save the machine

Inherited from Object:

Public Fields

Allocator* allocator

Public Methods

void addOption(const char* name, int size, void* ptr, const char* help="")

void addIOption(const char* name, int* ptr, int init_value, const char* help="")

void addROption(const char* name, real* ptr, real init_value, const char* help="")

void addBOption(const char* name, bool* ptr, bool init_value, const char* help="")

void addOOption(const char* name, Object** ptr, Object* init_value, const char* help="")

void setOption(const char* name, void* ptr)

void setIOption(const char* name, int option)

void setROption(const char* name, real option)

void setBOption(const char* name, bool option)

void setOOption(const char* name, Object* option)

void load(const char* filename)

void save(const char* filename)

void* operator new(size_t size, Allocator* allocator_=NULL)

void* operator new(size_t size, Allocator* allocator_, void* ptr_)

void operator delete(void* ptr)

Documentation

Trainer.

A trainer takes a Machine and is able to train this machine on a given dataset with the train() method. For each machine, it should exist a trainer which knows how to train this machine. Testing the machine is possible with the test() method.

Trainer(Machine* machine_)

virtual void train(DataSet* data_, MeasurerList* measurers) = 0

Train the machine. The Trainer has to call the measurers when it want.

virtual void test(MeasurerList* measurers)

Test the machine. This method call all the measurers, for all the examples of their associated dataset. It's already written...

static Allocator* extractMeasurers(MeasurerList* measurers, DataSet* train, DataSet*** datas, Measurer**** meas, int** n_meas, int* n_datas)

Make a table of measurers from a List.

Given a List of measurers, and, if you want, a train DataSet (else NULL)

• Returns all datasets associated to the measurers in datas. For i != j, (*datas)[i] != (*datas)[j]. Moreover, if train != NULL, (*datas)[0] = train.
• Returns the list of measurers associated to (*datas)[i] in (*meas)[i].
• Returns the number of measureurs associated to (*datas)[i] in (*n_meas)[i].
• Returns in *n_datas the number of datasets in *datas.

Returns an allocator to all the memory allocated by the function. You have to delete this allocator by yourself.

virtual void loadXFile(XFile* file)

By default, just load the machine

virtual void saveXFile(XFile* file)

By default, just save the machine

Direct child classes:

StochasticGradient
QCTrainer
NPTrainer
EMTrainer
Boosting
BayesClassifier
Bagging

Author:

Ronan Collobert (collober@idiap.ch)

Alphabetic index HTML hierarchy of classes or Java