Copyright	(c) Adam Scibior 2015-2020
License	MIT
Maintainer	leonhard.markert@tweag.io
Stability	experimental
Portability	GHC
Safe Haskell	Safe-Inferred
Language	Haskell2010

Control.Monad.Bayes.Sampler.Strict

Description

SamplerIO and SamplerST are instances of MonadDistribution. Apply a MonadFactor transformer to obtain a MonadMeasure that can execute probabilistic models.

Synopsis

data Sampler g m a
type SamplerIO = Sampler (IOGenM StdGen) IO
type SamplerST s = Sampler (STGenM StdGen s) (ST s)
sampleIO :: SamplerIO a -> IO a
sampleIOfixed :: SamplerIO a -> IO a
sampleWith :: Sampler g m a -> g -> m a
sampleSTfixed :: SamplerST s b -> ST s b
sampleMean :: [(Double, Log Double)] -> Double
sampler :: SamplerIO a -> IO a

Documentation

data Sampler g m a Source #

The sampling interpretation of a probabilistic program Here m is typically IO or ST

Instances

Instances details

MonadIO m => MonadIO (Sampler g m) Source #
Instance details Defined in Control.Monad.Bayes.Sampler.Strict Methods liftIO :: IO a -> Sampler g m a #
Applicative m => Applicative (Sampler g m) Source #
Instance details Defined in Control.Monad.Bayes.Sampler.Strict Methods pure :: a -> Sampler g m a # (<>) :: Sampler g m (a -> b) -> Sampler g m a -> Sampler g m b # liftA2 :: (a -> b -> c) -> Sampler g m a -> Sampler g m b -> Sampler g m c # (>) :: Sampler g m a -> Sampler g m b -> Sampler g m b # (<*) :: Sampler g m a -> Sampler g m b -> Sampler g m a #
Functor m => Functor (Sampler g m) Source #
Instance details Defined in Control.Monad.Bayes.Sampler.Strict Methods fmap :: (a -> b) -> Sampler g m a -> Sampler g m b # (<$) :: a -> Sampler g m b -> Sampler g m a #
Monad m => Monad (Sampler g m) Source #
Instance details Defined in Control.Monad.Bayes.Sampler.Strict Methods (>>=) :: Sampler g m a -> (a -> Sampler g m b) -> Sampler g m b # (>>) :: Sampler g m a -> Sampler g m b -> Sampler g m b # return :: a -> Sampler g m a #
StatefulGen g m => MonadDistribution (Sampler g m) Source #
Instance details Defined in Control.Monad.Bayes.Sampler.Strict Methods random :: Sampler g m Double Source # uniform :: Double -> Double -> Sampler g m Double Source # normal :: Double -> Double -> Sampler g m Double Source # gamma :: Double -> Double -> Sampler g m Double Source # beta :: Double -> Double -> Sampler g m Double Source # bernoulli :: Double -> Sampler g m Bool Source # categorical :: Vector v Double => v Double -> Sampler g m Int Source # logCategorical :: (Vector v (Log Double), Vector v Double) => v (Log Double) -> Sampler g m Int Source # uniformD :: [a] -> Sampler g m a Source # geometric :: Double -> Sampler g m Int Source # poisson :: Double -> Sampler g m Int Source # dirichlet :: Vector v Double => v Double -> Sampler g m (v Double) Source #

type SamplerIO = Sampler (IOGenM StdGen) IO Source #

convenient type synonym to show specializations of Sampler to particular pairs of monad and RNG

type SamplerST s = Sampler (STGenM StdGen s) (ST s) Source #

convenient type synonym to show specializations of Sampler to particular pairs of monad and RNG

sampleIO :: SamplerIO a -> IO a Source #

initialize random seed using system entropy, and sample

sampleIOfixed :: SamplerIO a -> IO a Source #

Run the sampler with a fixed random seed

sampleWith :: Sampler g m a -> g -> m a Source #

Sample with a random number generator of your choice e.g. the one from Random.

>>> import Control.Monad.Bayes.Class
>>> import System.Random.Stateful hiding (random)
>>> newIOGenM (mkStdGen 1729) >>= sampleWith random
4.690861245089605e-2

sampleSTfixed :: SamplerST s b -> ST s b Source #

Run the sampler with a fixed random seed

sampleMean :: [(Double, Log Double)] -> Double Source #

sampler :: SamplerIO a -> IO a Source #

initialize random seed using system entropy, and sample