cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType > Class Template Reference

The IterativePolicyEval class. More...

#include <iterative_policy_evaluation.h>

Inheritance diagram for cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >:

Collaboration diagram for cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >:

Public Types
typedef DPSolverBase< EnvType >::env_type	env_type
	env_type
typedef PolicyType	policy_type
	policy_type
Public Types inherited from cuberl::rl::algos::dp::DPSolverBase< EnvType >
typedef RLSolverBase< EnvType >::env_type	env_type
	The environment type the solver is using.
Public Types inherited from cuberl::rl::algos::RLSolverBase< EnvType >
typedef EnvType	env_type

Public Member Functions
	IterativePolicyEvalutationSolver (IterativePolicyEvalConfig config, policy_type &policy)
	IterativePolicyEval.
virtual void	actions_before_training_begins (env_type &env) override
	actions_before_training_begins. Execute any actions the algorithm needs before starting the iterations
virtual void	actions_after_training_ends (env_type &env) override
	actions_after_training_ends. Actions to execute after the training iterations have finisehd
virtual void	actions_before_episode_begins (env_type &, uint_t) override
	actions_before_training_episode
virtual void	actions_after_episode_ends (env_type &, uint_t, const EpisodeInfo &) override
	actions_after_training_episode
virtual EpisodeInfo	on_training_episode (env_type &env, uint_t episode_idx) override
	on_episode Do one on_episode of the algorithm
void	save (const std::string &filename) const
DynVec< real_t >	get_value_function () const
	value_function
policy_type	get_policy () const
	get_policy
void	update_policy (const policy_type &other)
	update_policy
Public Member Functions inherited from cuberl::rl::algos::dp::DPSolverBase< EnvType >
virtual	~DPSolverBase ()=default
	Destructor.
Public Member Functions inherited from cuberl::rl::algos::RLSolverBase< EnvType >
virtual	~RLSolverBase ()=default
	Destructor.

Protected Attributes
IterativePolicyEvalConfig	config_
DynVec< real_t >	v_
	v_
policy_type &	policy_
	policy_

Additional Inherited Members
Protected Member Functions inherited from cuberl::rl::algos::dp::DPSolverBase< EnvType >
	DPSolverBase ()=default
	DPAlgoBase.
Protected Member Functions inherited from cuberl::rl::algos::RLSolverBase< EnvType >
	RLSolverBase ()=default
	Constructor.

Detailed Description

template<typename EnvType, typename PolicyType>
class cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >

The IterativePolicyEval class.

Member Typedef Documentation

env_type

template<typename EnvType , typename PolicyType >

typedef DPSolverBase<EnvType>::env_type cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::env_type

env_type

policy_type

template<typename EnvType , typename PolicyType >

typedef PolicyType cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::policy_type

policy_type

Constructor & Destructor Documentation

IterativePolicyEvalutationSolver()

template<typename EnvType , typename PolicyType >

cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::IterativePolicyEvalutationSolver ( IterativePolicyEvalConfig  config, policy_type &  policy  )

explicit

IterativePolicyEval.

Member Function Documentation

actions_after_episode_ends()

template<typename EnvType , typename PolicyType >

virtual void cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::actions_after_episode_ends ( env_type &  , uint_t  , const EpisodeInfo &    )

inlineoverridevirtual

actions_after_training_episode

Reimplemented from cuberl::rl::algos::RLSolverBase< EnvType >.

actions_after_training_ends()

template<typename EnvType , typename PolicyType >

void cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::actions_after_training_ends ( env_type &  env )

overridevirtual

actions_after_training_ends. Actions to execute after the training iterations have finisehd

Implements cuberl::rl::algos::RLSolverBase< EnvType >.

actions_before_episode_begins()

template<typename EnvType , typename PolicyType >

virtual void cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::actions_before_episode_begins ( env_type &  , uint_t    )

inlineoverridevirtual

actions_before_training_episode

Reimplemented from cuberl::rl::algos::RLSolverBase< EnvType >.

actions_before_training_begins()

template<typename EnvType , typename PolicyType >

void cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::actions_before_training_begins ( env_type &  env )

overridevirtual

actions_before_training_begins. Execute any actions the algorithm needs before starting the iterations

Implements cuberl::rl::algos::RLSolverBase< EnvType >.

get_policy()

template<typename EnvType , typename PolicyType >

policy_type cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::get_policy ( ) const

inline

get_policy

Returns

get_value_function()

template<typename EnvType , typename PolicyType >

DynVec< real_t > cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::get_value_function ( ) const

inline

value_function

Returns

on_training_episode()

template<typename EnvType , typename PolicyType >

EpisodeInfo cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::on_training_episode ( env_type &  env, uint_t  episode_idx  )

overridevirtual

on_episode Do one on_episode of the algorithm

Implements cuberl::rl::algos::RLSolverBase< EnvType >.

save()

template<typename EnvType , typename PolicyType >

void cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::save

(

const std::string &

filename

)

const

update_policy()

template<typename EnvType , typename PolicyType >

void cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::update_policy ( const policy_type &  other )

inline

update_policy

Parameters

other

Member Data Documentation

config_

template<typename EnvType , typename PolicyType >

IterativePolicyEvalConfig cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::config_

protected

policy_

template<typename EnvType , typename PolicyType >

policy_type& cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::policy_

protected

policy_

v_

template<typename EnvType , typename PolicyType >

DynVec<real_t> cuberl::rl::algos::dp::IterativePolicyEvalutationSolver< EnvType, PolicyType >::v_

protected

v_

---

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

• libs/cuberl/include/cuberl/rl/algorithms/dp/iterative_policy_evaluation.h