maxent_toolbox

Maximum Entropy toolbox for MATLAB

View the Project on GitHub orimaoz/maxent_toolbox

maxent.getLogProbability

Description

Returns the log probability (in natural logarithm) of samples according to a trained maximum entropy model. The function accepts a model and an array of samples and returns the log probability for each sample according to the model. If the model has not been normalize, the results will also be un-normalized.

Usage

logprobs = maxent.getLogProbability(model, samples)
logprobs = maxent.getLogProbability(model, samples,Name,Value,...)

Arguments

Mandatory arguments

  • model - Maximum entropy model as returned by the trainModel function.
  • samples - Set of samples to get the probabilities of, in the format (ncells x nsamples).

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

  • normalize - force with or without normalization. If this value is true but the model is un-normalized, the function will issue a warning.

Output

  • logprobs - an (1xnsamples) vector of log probabilities in natural logarithm.