Benjamin Deonovic

For these kinds of situations I always find it easier to use what Brian calls the "stand alone functions" for the samplers. This is more of a manual way of...

Glad you figured out your problem @axsk. Let us know if you have any more questions. By the way did you see this issue: https://github.com/brian-j-smith/Mamba.jl/issues/42, there were lots of helpful...

It looks like `yx_mat` is a matrix, which could be why `yx_mat::AbstractArray{T}` or `yx_mat::Vector{T}` would throw errors. In this case you are trying to define a Matrix distribution (similar to...

1) Julia does differentiate between "vectors" and matrix of 1 column, although most functions will know how to handle both, e.g. note: ``` julia julia> [1,2,3] 3-element Array{Int64,1}: 1 2...

I'm glad you got your code working. Without being able to see any code I can't really give any helpful comments, but are you aware that not all your data...

I'm glad everything is working well. It's not really a trick but how the package was designed to work. Reading through the tutorial and the examples should be quite illuminating....

We are definitely trying to incorporate that. It is on the TODO list :) Currently some of the Mamba objects are hardcoded to be Float64s. If we can change to...

Seems like NUTS is getting stuck. Here is a reproducible script: ```julia using Mamba srand(123) ## set seed for reproducibility sigma_w = 1; N = 100; x = (1:N); line...

Changing initial values to something closer to the truth results in correct output: ```julia ## Initial Values inits = [ Dict{Symbol, Any}(:y => line[:y], :beta => [0.0, 1.0], :sigma_w =>...

The issue is because ``` julia data[:x] 2x30 Array{Float64,2}: ``` is a matrix of dimension 2 x 30. The way you coded up the stochastic node for `x` is ```...