updateSaveCobra: Update and save cobra

In SACOBRA: Self-Adjusting COBRA

Description

Helper for cobraPhaseII: make some assertion tests, update elements in object cobra, including data frames df and df2, and - if cobra$saveIntermediate==TRUE - save cobra in subdir results/. Most importantly cobra$xbest, cobra$fbest, cobra$ibest are updated. They characterize the best feasible point (least violating point if no feasible point was found so far) and influence the next starting point.

Usage

updateSaveCobra(
  cobra,
  ev1,
  subMin,
  sigmaD,
  penaF,
  gama,
  EPS,
  fitFuncPenalRBF,
  distRequirement,
  fitnessSurrogate = cobra$fitnessSurrogate
)

Arguments

cobra	an object of class COBRA, this is a (long) list containing all settings from cobraPhaseII
ev1	a list filled by calls to evalReal. We need here the elements xNew, feas, feasPred, feval, optimizerConvergence, optimizationTime, predY, predVal
subMin	see cobraPhaseII
sigmaD	see cobraPhaseII
penaF	see cobraPhaseII
gama	see cobraPhaseII
EPS	see cobraPhaseII
fitFuncPenalRBF	helper function from cobraPhaseII
distRequirement	helper function from cobraPhaseII
fitnessSurrogate	the model used for predSoluFunc

Details

Note: the elements A, Fres, Gres of cobra are set in updateInfoAndCounters, an internal function of cobraPhaseII.

Value

cobra, an object of class COBRA, enhanced here by the following elements (among others):

df	data frame with summary of the optimization run (see cobraPhaseII)
df2	data frame with additional summary information (see cobraPhaseII)
dftr	data frame with additional summary information for TR(see cobraPhaseII)
fbest	the best feasible objective value found
xbest	the point in input space yielding the best feasible objective value
ibest	the corresponding iteration number (row of cobra$df, of cobra$A)
fbestArray	vector of all fbest
xbestArray	vector of all xbest

See Also

cobraPhaseII, cobraInit

SACOBRA documentation built on March 26, 2020, 7:15 p.m.