toy_text_base.h

Go to the documentation of this file.

#ifndef TOY_TEXT_BASE_ENV_H
#define TOY_TEXT_BASE_ENV_H
 
#include "bitrl/bitrl_config.h"
#include "bitrl/bitrl_types.h"
#include "bitrl/envs/env_types.h"
#include "bitrl/envs/gymnasium/gymnasium_env_base.h"
#include "bitrl/extern/nlohmann/json/json.hpp"
#include "bitrl/network/rest_rl_env_client.h"
 
#include <any>
#include <string>
#include <tuple>
#include <unordered_map>
#include <vector>
 
#ifdef BITRL_DEBUG
#include <cassert>
#endif
 
namespace bitrl
{
namespace envs::gymnasium
{
 
 
template <typename TimeStepType, uint_t state_end, uint_t action_end>
class ToyTextEnvBase
    : public GymnasiumEnvBase<TimeStepType, ScalarDiscreteEnv<state_end, action_end, 0, 0>>
{
  public:
    typedef typename GymnasiumEnvBase<
        TimeStepType, ScalarDiscreteEnv<state_end, action_end, 0, 0>>::base_type base_type;
 
    typedef typename base_type::time_step_type time_step_type;
 
    typedef typename base_type::state_space_type state_space_type;
 
    typedef typename base_type::action_space_type action_space_type;
 
    typedef typename base_type::action_type action_type;
 
    typedef typename base_type::state_type state_type;
 
    typedef std::vector<std::tuple<real_t, uint_t, real_t, bool>> dynamics_t;
 
    virtual ~ToyTextEnvBase() = default;
 
    dynamics_t p(uint_t sidx, uint_t aidx) const;
 
    uint_t n_actions() const noexcept { return action_space_type::size; }
 
    uint_t n_states() const noexcept { return state_space_type::size; }
 
  protected:
    ToyTextEnvBase(network::RESTRLEnvClient &api_server, const std::string &name);
 
    ToyTextEnvBase(const ToyTextEnvBase &other);
 
    virtual dynamics_t build_dynamics_from_response_(const nlohmann::json &response) const;
};
 
template <typename TimeStepType, uint_t state_end, uint_t action_end>
ToyTextEnvBase<TimeStepType, state_end, action_end>::ToyTextEnvBase(
    network::RESTRLEnvClient &api_server, const std::string &name)
    : GymnasiumEnvBase<TimeStepType, ScalarDiscreteEnv<state_end, action_end>>(api_server, name)
{
}
 
template <typename TimeStepType, uint_t state_end, uint_t action_end>
ToyTextEnvBase<TimeStepType, state_end, action_end>::ToyTextEnvBase(
    const ToyTextEnvBase<TimeStepType, state_end, action_end> &other)
    : GymnasiumEnvBase<TimeStepType, ScalarDiscreteEnv<state_end, action_end>>(other)
{
}
 
template <typename TimeStepType, uint_t state_end, uint_t action_end>
typename ToyTextEnvBase<TimeStepType, state_end, action_end>::dynamics_t
ToyTextEnvBase<TimeStepType, state_end, action_end>::p(uint_t sidx, uint_t aidx) const
{
 
#ifdef BITRL_DEBUG
    assert(this->is_created() && "Environment has not been created");
#endif
 
    auto response = this->get_api_server().dynamics(this->env_name(), this->idx(), sidx, aidx);
    return build_dynamics_from_response_(response);
}
 
template <typename TimeStepType, uint_t state_end, uint_t action_end>
typename ToyTextEnvBase<TimeStepType, state_end, action_end>::dynamics_t
ToyTextEnvBase<TimeStepType, state_end, action_end>::build_dynamics_from_response_(
    const nlohmann::json &response) const
{
    auto dynamics = response["dynamics"];
    return dynamics;
}
 
} // namespace envs::gymnasium
} // namespace bitrl
#endif