nArmedBanditsSimpleQlearning.hpp

Go to the documentation of this file.

#ifndef SRC_APPLICATION_NARMEDBANDITSSIMPLEQLEARNING_HPP_
#define SRC_APPLICATION_NARMEDBANDITSSIMPLEQLEARNING_HPP_

#include <vector>

#include <types/MatrixTypes.hpp>

#include <opengl/application/OpenGLApplication.hpp>
#include <opengl/visualization/WindowCollectorChart.hpp>
using namespace mic::opengl::visualization;

namespace mic {
namespace application {

class TestApp: public mic::opengl::application::OpenGLApplication {
public:
    TestApp(std::string node_name_ = "application");

    virtual ~TestApp();

protected:
    virtual void initializePropertyDependentVariables();

    virtual void initialize(int argc, char* argv[]);

    virtual bool performSingleStep();

private:

    WindowCollectorChart<float>* w_reward;

    mic::utils::DataCollectorPtr<std::string, float> reward_collector_ptr;

    mic::types::VectorXf arms;

    mic::types::VectorXf action_values;

    mic::types::VectorXi action_counts;

    mic::configuration::Property<size_t> number_of_bandits;

    mic::configuration::Property<double> epsilon;

    mic::configuration::Property<std::string> statistics_filename;

    size_t best_arm;

    float best_arm_prob;

    short calculateReward(float prob_);

    size_t selectBestArm();

};

} /* namespace application */
} /* namespace mic */

#endif /* SRC_APPLICATION_NARMEDBANDITSSIMPLEQLEARNING_HPP_ */