Brian J Smith

Hi Feynman, Nice work tracking down the issue and creating a pull request to add an ``insupport`` method to the Distribution package's MixtureModel type. This is the first time I...

I have reservations about code that attempts to guess at variable types (continuous, discrete, or binary), similar to those expressed in #112. Such guesses could vary across different runs of...

As noted, corrected files are in the Mamba 0.12 documentation. The "latest" documentation link will be pointed to version 0.12 as soon as it is publicly released, for which we...

Typically, sampled stochastic nodes are given different initial values for each chain that is run; whereas, input (data and prior) values are constant across all chains. Thus, initial values and...

Sure. Probably the most straightforward approach would be to use one of the stand-alone samplers, like that for [slice sampling](http://mambajl.readthedocs.org/en/latest/samplers/slice.html), as follows. P.S. The code below requires julia 0.4. ```...

A Model() approach can definitely be used: - `y` would be a stochastic node having a Normal distribution with mean `f(x)` and standard deviation 1.0 - the likelihood. - `x`...

I now have an implementation, running locally, that allows functions in the specification of Stochastic and Logical nodes. A preview of the syntax is shown below for part of the...

Function syntax for the specification of Model nodes is officially supported now in the [latest release](http://mambajl.readthedocs.org/en/latest/) of Mamba (0.7.0), along with full documentation and updated [tutorial](http://mambajl.readthedocs.org/en/latest/tutorial.html) and [examples](http://mambajl.readthedocs.org/en/latest/examples.html).

Hi @mikejacktzen, Congratulations on the progress! Have you tried your modified logpdf approach in recent versions of julia (0.4) and Mamba (0.7)? The package is a little more stringent now...

Awesome! So glad to hear that, and thanks for the updates along the way.