[Master Index] [Index for export/lyngby]



Function Synopsis

[TrainIndices, ValIndices] = lyngby_nn_setsplit(arg1, arg2, ...

Help text

 lyngby_nn_setsplit   - Split the data into training and validation set

       function [TrainIndices, ValIndices] =
           lyngby_nn_setsplit('PropertyName', 'PropertyValue')

       Input:  Property:
                  'Type'           [ {Blocked} | Interlaced ]
                  'Run'            Run definition
                  'Paradigm'       Paradigm
                  'TrainFraction'  { 0.5 } The part of the total set
                                   that should be the training set  

       Output: TrainIndices   Index for the training set
               ValIndices     Index for the validation set

       This function will split the data into a training set and a
       validation set. If 'Type' is set to 'Blocked' and if the 'Run'
       information is provided it will use this to split the data:
       The scans in a single run will not be split in two groups. 
       If 'Type' is set to 'Interlaced' every 1/TrainFraction'th is
       put in the training set and the rest in the validation set

       The default is that the first scans (until 'TrainFraction') is
       returned as the training set and the reminder is returned as
       the validation set.

       See also: lyngby_runinfo

Cross-Reference Information

This function calls
This function is called by

Produced by mat2html on Wed Jul 29 15:43:40 2009
Cross-Directory links are: OFF