A hidden Markov model (HMM) is a statistical tool for modelling sequential processes. The model consists of a Markov chain of discrete unobserved variables, each of which emits an observed continuous variable. The probabilistic dependencies between each of the variables in the chain is governed by a set of model parameters. Below is a graphical model of an HMM:

This project provides a C# definition for a HMM using the Infer.NET framework. This code makes it easy to run approximate Bayesian inference over both the model parameters and states of a HMM.

Special thanks go to Microsoft Research for adding support for chain models, and to Matteo Venanzi for his expertise in increasing the efficiency of the model.

• .NET Framework 4.5

• Infer.NET 2.6