Math.HiddenMarkovModel.Example.Circle

Warning: do not import that module, it is only intended for demonstration

Example of an HMM with continuous emissions with two-dimensional observations. We train a model to accept a parametric curve of a circle with a certain speed. This is like Math.HiddenMarkovModel.Example.SineWave but in two dimensions.

The four hidden states correspond to the four quadrants.

Documentation

hmm :: Gaussian Double

circleLabeled :: T [] (State, Vector Double)

circle :: T [] (Vector Double)

revealed :: T [] State

reconstructDistribution :: Gaussian Double

Sample multivariate normal distribution and reconstruct it from the samples. You should obtain the same parameters.

hmmTrainedSupervised :: Gaussian Double

hmmTrainedUnsupervised :: Gaussian Double

hmmIterativelyTrained :: Gaussian Double