Safe Haskell	None
Language	GHC2021

RL.ReplayBuffer

Documentation

newtype RPState tr tr' slc s f h Source #

Constructors

Instances details

(Show tr, Show tr', Show slc, Show s, Show f, Show h) => Show (RPState tr tr' slc s f h) Source #
Instance details Defined in RL.ReplayBuffer Methods showsPrec :: Int -> RPState tr tr' slc s f h -> ShowS # show :: RPState tr tr' slc s f h -> String # showList :: [RPState tr tr' slc s f h] -> ShowS #

newtype RPAction slc tr s f h Source #

Constructors

RPAction (Action slc tr s f h)

Instances details

(Show slc, Show tr, Show s, Show f, Show h) => Show (RPAction slc tr s f h) Source #
Instance details Defined in RL.ReplayBuffer Methods showsPrec :: Int -> RPAction slc tr s f h -> ShowS # show :: RPAction slc tr s f h -> String # showList :: [RPAction slc tr s f h] -> ShowS #

data ReplayStep (dev :: (DeviceType, Nat)) tr tr' slc s f h Source #

Constructors

ReplayStep
Fields replayState :: !(RPState tr tr' slc s f h) replayAction :: !(RPAction slc tr s f h) replayStep :: !(QEncoding dev ('[] :: [Nat])) replayNextState :: !(Maybe (RPState tr tr' slc s f h)) replayNextSteps :: ![QEncoding dev ('[] :: [Nat])] replayReward :: !QType

Instances details

(Show slc, Show s, Show f, Show h, Show tr, Show tr') => Show (ReplayStep dev tr tr' slc s f h) Source #
Instance details Defined in RL.ReplayBuffer Methods showsPrec :: Int -> ReplayStep dev tr tr' slc s f h -> ShowS # show :: ReplayStep dev tr tr' slc s f h -> String # showList :: [ReplayStep dev tr tr' slc s f h] -> ShowS #

data ReplayBuffer (dev :: (DeviceType, Nat)) tr tr' slc s f h Source #

Constructors

ReplayBuffer !Int ![ReplayStep dev tr tr' slc s f h]

Instances details

(Show slc, Show s, Show f, Show h, Show tr, Show tr') => Show (ReplayBuffer dev tr tr' slc s f h) Source #
Instance details Defined in RL.ReplayBuffer Methods showsPrec :: Int -> ReplayBuffer dev tr tr' slc s f h -> ShowS # show :: ReplayBuffer dev tr tr' slc s f h -> String # showList :: [ReplayBuffer dev tr tr' slc s f h] -> ShowS #

mkReplayBuffer :: forall (dev :: (DeviceType, Nat)) tr tr' slc s f h. Int -> ReplayBuffer dev tr tr' slc s f h Source #

seedReplayBuffer :: forall (dev :: (DeviceType, Nat)) tr tr' slc s f h. Int -> [ReplayStep dev tr tr' slc s f h] -> ReplayBuffer dev tr tr' slc s f h Source #

pushStep :: forall (dev :: (DeviceType, Nat)) tr tr' slc s f h. ReplayBuffer dev tr tr' slc s f h -> ReplayStep dev tr tr' slc s f h -> ReplayBuffer dev tr tr' slc s f h Source #

sampleSteps :: forall (dev :: (DeviceType, Nat)) tr tr' slc s f h. ReplayBuffer dev tr tr' slc s f h -> Int -> IO [ReplayStep dev tr tr' slc s f h] Source #