package prbnmcn-dagger

You can search for identifiers within the package.

in-package search v0.2.0

Probabilistic programming library

Install

Dune Dependency

github.com Readme Changelog MIT License Edit opam file Versions (6)

Authors

I

Ilias Garnier

Maintainers

I

igarnier@protonmail.com

Sources

0.0.1.tar.gz

md5=304d489ed3e9605fcd7fe3f86c2e9dca

sha512=f8f632e381c6b1d111ee8a03e142994f2c59ddabdbb5c3fef86c7c41a95ad4835c4ea4a893fa18e2f6522b92be8fd9b499c12b21a008bc647a85277ba5bccc5e

doc/prbnmcn-dagger/Dagger/Dist/index.html

Module `Dagger.Dist`Source

Distributions

A distribution must be sampleable and must be associated to a density w.r.t. some base measure (which is kept implicit).

Sourcetype 'a dist = private {

sample : RNG.t -> 'a;
ll : 'a -> Log_space.t;

}

'a dist is the type of distributions on the type 'a.

Sourcetype 'a kernel = private {

start : 'a;
sample : 'a -> RNG.t -> 'a;
ll : 'a -> 'a -> Log_space.t;

}

'a kernel is the type of kernels (ie random walks) on the type 'a.

Sourcetype 'a t = private

| Stateless of 'a dist
| Kernel of 'a kernel

A "distribution" as understood by this library is either a kernel or a plain distribution.

Sourceval stateless : (RNG.t -> 'a) -> ('a -> Log_space.t) -> 'a t

Create a distribution from a sampler and a log-density.

Sourceval kernel : 'a -> ('a -> RNG.t -> 'a) -> ('a -> 'a -> Log_space.t) -> 'a t

Create a kernel from an initial point, a kernel and its associated log-density.

Helpers for parameterized distributions and kernels.

Sourceval dist0 : (RNG.t -> 'a) -> ('a -> Log_space.t) -> 'a t

Sourceval dist1 : ('a -> RNG.t -> 'b) -> ('a -> 'b -> Log_space.t) -> 'a -> 'b t

Source

val dist2 : 
  ('a -> 'b -> RNG.t -> 'c) ->
  ('a -> 'b -> 'c -> Log_space.t) ->
  'a ->
  'b ->
  'c t

Source

val kernel1 : 
  ('a -> 'b -> RNG.t -> 'b) ->
  ('a -> 'b -> 'b -> Log_space.t) ->
  'b ->
  'a ->
  'b t

Combinators on distributions and kernels.

Sourceval iid : int -> 'a t -> 'a array t

iid n dist constructs the n-fold iid distribution where each component is distributed according to dist.

Sourceval conv : ('a -> 'b) -> ('b -> 'a) -> 'a t -> 'b t

conv f g dist transports a distribution along a pair of maps f,g. f,g must both be total. They need not be bijections, though not respecting this additional constraint might cause unspecified behaviours for some inference algorithms..