Control.Monad.Bayes.Weighted

Weighted is an instance of MonadCond. Apply a MonadSample transformer to
obtain a MonadInfer that can execute probabilistic models.

Reexports

Definitions

record Weighted : (Type -> Type) -> Type -> Type

  Execute the program using the prior distribution, while accumulating likelihood.

Totality: total
Visibility: public export
Constructor: 

MkWeighted : StateT (Log Double) m a -> Weighted m a

Projection: 

.runWeighted' : Weighted m a -> StateT (Log Double) m a

Hints:

Monad m => Applicative (Weighted m)

Functor m => Functor (Weighted m)

Monad m => Monad (Weighted m)

Monad m => MonadCond (Weighted m)

MonadSample m => MonadInfer (Weighted m)

MonadSample m => MonadSample (Weighted m)

MonadTrans Weighted

.runWeighted' : Weighted m a -> StateT (Log Double) m a

Visibility: public export

runWeighted' : Weighted m a -> StateT (Log Double) m a

Visibility: public export

runWeighted : Weighted m a -> m (Log Double, a)

  Obtain an explicit value of the likelihood for a given value

Visibility: public export

prior : Functor m => Weighted m a -> m a

  Compute the sample and discard the weight.
  This operation introduces bias.

Visibility: export

extractWeight : Functor m => Weighted m a -> m (Log Double)

  Compute the weight and discard the sample.

Visibility: export

withWeight : Monad m => m (Log Double, a) -> Weighted m a

  Embed a random variable with explicitly given likelihood

Visibility: export

flatten : Monad m => Weighted (Weighted m) a -> Weighted m a

  Combine weights from two different levels.

Visibility: export

applyWeight : MonadCond m => Weighted m a -> m a

  Use the weight as a factor in the transformed monad.

Visibility: export

hoist : (m x -> n x) -> Weighted m a -> Weighted n a

  Apply a transformation to the transformed monad.

Visibility: export