Main Content


Remove members of compact regression ensemble


cens1 = removeLearners(cens,idx)


cens1 = removeLearners(cens,idx) creates a compact regression ensemble identical to cens only without the ensemble members in the idx vector.

Input Arguments


Compact regression ensemble, constructed with compact.


Vector of positive integers with entries from 1 to cens.NumTrained, where cens.NumTrained is the number of members in cens. cens1 contains the members of cens except those with indices in idx.

Typically, you set idx = j:cens.NumTrained for some positive integer j.

Output Arguments


Compact regression ensemble, identical to cens except cens1 does not contain members of cens with indices in idx.


expand all

Create a compact regression ensemble. Compact it further by removing members of the ensemble.

Load the carsmall data set and select Weight and Cylinders as predictors.

load carsmall
X = [Weight Cylinders];

Train a regression ensemble using LSBoost. Specify tree stumps as the weak learners.

t = templateTree('MaxNumSplits',1);
ens = fitrensemble(X,MPG,'Method','LSBoost','Learners',t,...

Create a compact classification ensemble cens from ens.

cens = compact(ens);

Remove the last 50 members of the ensemble.

idx = cens.NumTrained-49:cens.NumTrained;
cens1 = removeLearners(cens,idx);


  • Typically, set cens1 equal to cens to retain just one ensemble.

  • Removing learners reduces the memory used by the ensemble and speeds up its predictions.

Extended Capabilities