Probabilistic programming via source rewriting

WIP successor to Soss.jl

A Julia implementation of Bayesian linear regression using marginal likelihood for hyperparameter optimization, as presented in Chris Bishop's book.

KeywordCalls makes it easy to define a method taking a NamedTuple considered as a an unordered collection of bound variables. The required redirection is done at compile time, so there's no runtime overhead.

SossMLJ makes it easy to build MLJ machines from user-defined models from the Soss probabilistic programming language

-

-

-

-

-

-

A constant RNG, for cases when you need high efficiency and don't care about randomness

-

-