FSMLR

Fast Stepwise Stagewise Multivariateple Linear Regressio= n (FSMLR) (previous name Fast Stepwise Multiple Linear Regress= ion) is a procedure for stepwise stagewise building of linear regressi= on models by means of greedy descriptor selection.
It can be viewed as= a special case of the additive regression procedure (regression boosting) = specially designed to be compatible with the three-set approach based on th= e use of three different sets for learning: training set, internal tuning v= alidation set and external test set.

The main configurable parameters are:

shrinkage =E2=80=93 its decrease leads to the decrease of generalizatio= n error and increase of the number of required iterations, and
the relative size of the internal tuning validation set used for stoppi= ng descriptor selection procedure.

Reference

Zhokhova N, Baskin I, Palyulin V, Zefirov A, Zefirov N. Fragmental desc= riptors with labeled atoms and their application in QSAR/QSPR studies. Dokl= ady Chemistry. 2007 Dec 1;417(2):282-284.

FSMLR

Reference Zhokhova N, Baskin I, Palyulin V, Zefirov A, Zefirov N. Fragmental desc= riptors with labeled atoms and their application in QSAR/QSPR studies. Dokl= ady Chemistry. 2007 Dec 1;417(2):282-284.

Reference

Zhokhova N, Baskin I, Palyulin V, Zefirov A, Zefirov N. Fragmental desc= riptors with labeled atoms and their application in QSAR/QSPR studies. Dokl= ady Chemistry. 2007 Dec 1;417(2):282-284.