Self-Adjusting COBRA

[Up] [Top]

Documentation for package ‘SACOBRA’ version 1.2

Help Pages

SACOBRA-package Self-adjusting Constrained Optimization with RBF Surrogates
cobraInit Initial phase for SACOBRA optimizer
cobraPhaseI Find a feasible solution.
cobraPhaseII Improve the feasible solution by searching new infill points
COP Constraint Optimization Problem Benchmark (G Function Suite)
defaultCA Default settings for online whitening functionality
defaultDebugRBF Default settings for debug visualization RBF (only for 'd==2')
defaultEquMu Default settings for equality handling mechanism
defaultMS Default settings for the model-selection part of SACOBRA.
defaultRI Default settings for 'repairInfeasRI2' and 'repairChootinan'.
defaultSAC Default settings for the SACOBRA part of SACOBRA.
defaultTR Default settings for the trust-region part of COBRA.
distLine Euclidean distance of 'x' to all 'xp'
DRCL Distance Requirement Cycle, long version
DRCS Distance Requirement Cycle, short version
evalReal Evaluate new iterate on real function(s)
forwardRescale Forward Rescaling
getFbest Return best objective function value
getXbest Return best feasible solution in original space
intern.archive.env Archiving Environment
interpRBF Apply the trained cubic, MQ or Gaussian RBF interpolation to new data for d>1.
inverseRescale Inverse Rescaling
multiCOBRA Perform multiple COBRA runs
multiRunPlot Plot the results from multipe COBRA runs.
multiRunPlot_2 Plot the results from multipe COBRA runs.
plog Monotonic transform
plogReverse Inverse of 'plog'
predict.RBFinter Apply cubic or Gaussian or MQ RBF interpolation
repairChootinan Repair an infeasible solution with the method of Chootinan.
repairInfeasRI2 Repair an infeasible solution with the method RI2
rescaleWrapper Return a rescaled function
SACOBRA Self-adjusting Constrained Optimization with RBF Surrogates
setOpts Merge the options from a partial list and the default list
startCobra Start COBRA (constraint-based optimization) phase I and/or phase II
trainCubicRBF Fit cubic RBF interpolation to training data X for d>1.
trainGaussRBF Fit Gaussian RBF model to training data for d>1.
trainMQRBF Fit multiquadric RBF model to training data for d>1.
trustRegion Performs trust region refinement