cens = compact(ens)
cens = compact(ens) creates a compact version of ens. You can predict classifications using cens exactly as you can using ens. However, since cens does not contain training data, you cannot perform some actions, such as cross validation.
A classification ensemble created with fitensemble.
A compact classification ensemble. cens has class CompactClassificationEnsemble.
Compare the size of a classification ensemble for Fisher's iris data to the compact version of the ensemble:
load fisheriris ens = fitensemble(meas,species,'AdaBoostM2',100,'Tree'); cens = compact(ens); b = whos('ens'); % b.bytes = size of ens c = whos('cens'); % c.bytes = size of ens [b.bytes c.bytes] % shows cens uses less memory ans = 571727 532476