maxent_toolbox

Maximum Entropy toolbox for MATLAB

View the Project on GitHub orimaoz/maxent_toolbox

maxent.getMarginals

Description

Returns the marginals of a maximum entropy model, by either exhaustive computation (for small models) or MCMC sampling (for big models).

Usage

marginals = maxent.getMarginals(model)
marginals = maxent.getMarginals(model,Name,Value,...)

Arguments

Mandatory arguments

  • model - Maximum entropy model as returned by the trainModel function.

Optional arguments (in the form of name,value pairs)

  • force_exhaustive - force exhaustive computation. This could incur a very long runtime for big models.
  • force_mcmc - force MCMC (sampling-based) computation, which returns only an approximate result.
  • nsamples - number of empirical samples used to compute the marginals for the MCMC-based method.

Output

  • marginals - an (1xnfactors) vector of marginals.