mutable struct Agent
learner::AbstractReinforcementLearner
policy::AbstractPolicy
callback::AbstractCallback
TabularReinforcementLearning.Agent
— Method.Agent(learner; policy = EpsilonGreedyPolicy(.1), callback = NoCallback())
TabularReinforcementLearning.Agent
— Method.Agent(learner::NstepLearner; policy = EpsilonGreedyPolicy(.1), callback = NoCallback())
Replaces policy
with SoftmaxPolicy1 for baselearner of type AbstractPolicyGradient.
TabularReinforcementLearning.Agent
— Method.Agent(learner::AbstractPolicyGradient; policy = SoftmaxPolicy1(), callback = NoCallback())
mutable struct RLSetup
agent::Agent
environment
metric::AbstractEvaluationMetrics
stoppingcriterion::StoppingCriterion
TabularReinforcementLearning.learn!
— Method.learn!(learner, policy, callback, environment, metric, stoppingcriterion)
TabularReinforcementLearning.learn!
— Method.learn!(agent::Agent, environment, metric, stoppingcriterion)
TabularReinforcementLearning.learn!
— Method.learn!(x::RLSetup)
TabularReinforcementLearning.run!
— Function.run!(learner, policy, callback, environment, metric, stoppingcriterion)
TabularReinforcementLearning.run!
— Method.run!(agent::Agent, environment, metric, stoppingcriterion)
TabularReinforcementLearning.run!
— Method.run!(x::RLSetup)