metaheurhyper-class


An S4 class to represent hyperparameter optimization results


Automation of preprocessing often requires computationally costly
preprocessing combinations. This package helps to find near-best combinations
faster. Metaheuristics supported are taboo search, simulated annealing, reheating
and late acceptance. Start conditions include random and grid starts. End conditions
include all iteration rounds completed, objective threshold reached and convergence.
Metaheuristics, start and end conditions can be hybridized and hyperparameters optimized.
Parallel computations are supported. The package is intended to be used with package
'preprocomb' and takes its 'GridClass' object as input.

Markus Vattulainen

metaheur

Metaheuristic Optimization Framework for Preprocessing
Combinations

metaheurhyper-class function

 <dl>
<dt><code>res</code></dt><dd>(matrix) of classification accuracy results, rows for iterations and columns for parameter combinations</dd> <dt><code>namegrid</code></dt><dd>(list) parameter combinations</dd> <dt><code>baseheur</code></dt><dd>(list) metaheur objects</dd> <dt><code>walltime</code></dt><dd>(integer) execution wall-clock time in minutes</dd> </dl>

metaheurhyper-class: An S4 class to represent hyperparameter optimization results

Description

Arguments

Slots