bitrl & cuberl Documentation
Simulation engine for reinforcement learning agents
Loading...
Searching...
No Matches
example_15.cpp File Reference
#include "cubeai/base/cubeai_types.h"
#include "cubeai/rl/algorithms/algorithm_base.h"
#include "cubeai/rl/policies/epsilon_greedy_policy.h"
#include "cubeai/rl/utilities/vector_experience_buffer.h"
#include "cubeai/utils/array_utils.h"
#include "gymfcpp/gymfcpp_types.h"
#include "gymfcpp/mountain_car_env.h"
#include "gymfcpp/time_step.h"
#include <boost/python.hpp>
#include <vector>
#include <iostream>
#include <unordered_map>
#include <utility>
Include dependency graph for example_15.cpp:

Classes

struct  example::Transition
 
struct  example::Policy
 
class  example::ApproxMC< Env >
 

Namespaces

namespace  example
 

Typedefs

typedef gymfcpp::TimeStep< uint_texample::time_step_t
 
typedef std::pair< uint_t, uint_texample::state_type
 

Functions

template<typename ItrTp , typename StateTp >
ItrTp example::is_state_included (ItrTp begin, ItrTp end, const StateTp &state)
 
template<typename BinType >
std::pair< uint_t, uint_texample::get_aggregated_state (const std::pair< real_t, real_t > &obs, const BinType &pos_bins, const BinType &vel_bins)
 
int main ()
 

Variables

const real_t example::GAMMA = 1.0
 
const uint_t example::N_EPISODES = 20000
 
const uint_t example::N_ITRS_PER_EPISODE = 2000
 
const real_t example::TOL = 1.0e-8
 
auto example::pos_bins = std::vector<real_t>({-1.2, -0.95714286, -0.71428571, -0.47142857, -0.22857143, 0.01428571, 0.25714286, 0.5})
 
auto example::vel_bins = std::vector<real_t>({-0.07, -0.05, -0.03, -0.01, 0.01, 0.03, 0.05, 0.07})
 

Function Documentation

◆ main()

int main ( )