Model selection facility.
Select a model among multiple models (i.e., a parametric model, parametrized by a set of hyperparamenters).
Methods
max_log_marginal_likelihood(hyp_initial_guess) | Set up the optimization problem in order to maximize the log_marginal_likelihood. |
solve([problem]) | Solve the maximization problem, check outcome and collect results. |
TODO:
Methods
max_log_marginal_likelihood(hyp_initial_guess) | Set up the optimization problem in order to maximize the log_marginal_likelihood. |
solve([problem]) | Solve the maximization problem, check outcome and collect results. |
Set up the optimization problem in order to maximize the log_marginal_likelihood.
Parameters: | parametric_model : Classifier
hyp_initial_guess : numpy.ndarray
optimization_algorithm : string
ftol : float
fixedHypers : numpy.ndarray (boolean array)
|
---|
Notes
The maximization of log_marginal_likelihood is a non-linear optimization problem (NLP). This fact is confirmed by Dmitrey, author of OpenOpt.
Solve the maximization problem, check outcome and collect results.