I’ve now cleaned up (in befb0f3cca0c212e368497e86f030aa96355be18) the Reader and Writer interfaces and added it to Statistics.GModeling.Gibbs. I’ve removed references to Support and simply parameterized using a key type k and value type v. > data Reader k v = Reader … Continue reading
I am putting together what I have so far in a repository here. So far, (133e22dc979d988706aafe52a346cee004f70ca5) it contains Statistics.GModeling.DSL Statistics.GModeling.Models.HMM Statistics.GModeling.Models.LDA Statistics.GModeling.Models.FixedTreeLDA Will continue building the pieces in upcoming posts.
I think most have heard something like you only need suprisingly few people in a room before two people in the room end up sharing a birthday. But I never bothered to work it out. Let me do that. First, … Continue reading
A quick aside. I was thinking about how response variables are attached to generative models. For instance, if we want to say have binary classification on documents we would normally 1) take the dot product the topic vector with a … Continue reading
The next step is for the library to have access to the latent variable data. I also don’t want the library to decide how to store the data because the user will have a much better idea of what is … Continue reading
In this post, I write some functions to interpret the DSL. Specifically, I present some functions to figure out the children and parents of a node and discover what the prior, observed, and latent variables are. > import Control.Monad (msum) … Continue reading
In the previous post I attempted to introduce a DSL for probabilistic models inspired by the plate notation. Let’s try to see if we can define LDA with it. > data LDALabels = Alpha | Beta | Topics | Topic … Continue reading