| PRTools contents |
FEATSELLR
[W,RES] = FEATSELLR(A,CRIT,K,L,R,T,FID)
| Input | |
| A | Dataset |
| CRIT | String name of the criterion or untrained mapping (optional; default: 'NN', i.e. 1-Nearest Neighbor error) |
| K | Number of features to select (optional; default: return optimally ordered set of all features) |
| L | Number of features to select at a time (plus-L, default: 1), L ~= R |
| R | Number of features to deselect at a time (takeaway-R, default: 0) |
| T | Tuning set (optional) |
| N | Number of cross-validations (optional) |
| FID | File ID to write progress to (default [], see PRPROGRESS) |
| Output | |
| W | Output feature selection mapping |
| RES | Matrix with step-by-step results of the selection |
Floating selection of K features using the dataset A, by iteratively selecting L optimal features and deselecting R. Starts from the full set of features when L < R, otherwise from the empty set. CRIT sets the criterion used by the feature evaluation routine FEATEVAL. If the dataset T is given, it is used as a tuning set for FEATEVAL. Alternatively a number of cross-validations N may be supplied. For K = 0, the optimal feature set (maximum value of FEATEVAL) is returned. The result W can be used for selecting features by B*W. In this case, features are ranked optimally. The selected features are stored in W.DATA and can be found by +W. In R, the search is reported step by step as
RES(:,1) : number of features
RES(:,2) : criterion value
RES(:,3:3+max(L,R)) : added / deleted features
mappings, datasets, feateval, featsel, featselo, featselb, featself, featseli, featselp, featselm, prprogress,
| PRTools contents |