probability

Probabilistic Functional Programming

http://www.haskell.org/haskellwiki/Probabilistic_Functional_Programming

Version on this page:0.2.8
LTS Haskell 23.1:0.2.8@rev:1
Stackage Nightly 2024-12-22:0.2.8@rev:1
Latest on Hackage:0.2.8@rev:1

See all snapshots probability appears in

BSD-3-Clause licensed by Martin Erwig, Steve Kollmansberger
Maintained by Henning Thielemann
This version can be pinned in stack with:probability-0.2.8@sha256:d3bc4fe1e4eaa4c27d0e094b09ae25d2ba4a35afc5b5ca03d91e25e1f631fdac,2720

Module documentation for 0.2.8

The Library allows exact computation with discrete random variables in terms of their distributions by using a monad. The monad is similar to the List monad for non-deterministic computations, but extends the List monad by a measure of probability. Small interface to R plotting.

Changes

Change log for the probability package

0.2.8

  • Numeric.Probability.Either -> Control.Monad.Trans.Except Few functions exposed EitherT in their signatures, but Numeric.Probability.Either was private. Thus its removal should not be noticed by library users.

0.2.7

  • support random-1.2

0.2.6

  • instance Monad Distribution.T: Remove definition of fail. This turns calls to fail into errors for GHCs prior to the “Monad Fail Proposal”. Formerly it was an empty list, but this was bad since the probabilities in an empty list sum up to zero not one, thus breaking the invariant. Beginning with GHC-8.8 and the “Monad Fail Proposal” you can no longer accidentally call fail, since Distribution.T is not an instance of MonadFail.

  • instance Monad Probability.EitherT: Define MonadFail instance for GHC>=8.8.1.