A Julia wrapper for the Smile C++ Structural Modeling, Inference, and Learning Engine. This grants access to a wide variety of graphical decision-theoretical methods, such as Bayesian Networks and influence diagrams.

Smile.jl is no longer needed as of Julia 0.5. Julians can use the Cxx library to call the Smile C++ .so files directly. There is also the BayesNets package which provides native Julia implementations.

Library documentation is available here.

Smile can be installed through the Julia package manager (version 0.3 required)

julia> Pkg.add("Smile")

Instructions are available as well for those who wish to compile the C++ wrapper themselves.

The Decision Systems Laboratory asks that all publications of research in which SMILE is used contain an explicit acknowledgement to that effect. A simple example is: "The models described in this paper were created using the SMILE reasoning engine for graphical probabilisitc models developed by the Decision Systems Laboratory at the University of Pittsburgh and available at http://genie.sis.pitt.edu/."