You can indeed use Infer.NET to visualise a graphical model. However, our framework operates on factor graphs instead of Bayes nets. Therefore, you'll need to convert your model before running inference. Once the factor graph is
implemented in terms of the Infer.NET modelling API, you can visualise it by setting the inference engine
*ShowFactorGraph* property to *true*. An example of a displayed factor graph can be found in our
User Guide.

Factor graphs are very powerful, and you might want to consider them for your application. A good introduction to the concept was given by Chris Bishop here (jump
to *'Probabilistic Graphical Models'* on the right, or even more concretely to
*'From Directed Graph to Factor Graph'*). An extended talk on graphical models in general can be found
here.