Home > SFMI > DefaultTuningOptions.m

DefaultTuningOptions

PURPOSE ^

Create a vector containing the default values for tuning preferences

SYNOPSIS ^

function TuningOptions=DefaultTuningOptions(TuningMethod)

DESCRIPTION ^

 Create a vector containing the default values for tuning preferences

 function TuningOptions=DefaultTuningOptions

 Create a vector containing the default values for system tuning
 preferences
 -------------------------------------------------------------------------
 Parameters
 -------------------------------------------------------------------------
 TuningMethod - string containing the name of the selected tuning method
 -------------------------------------------------------------------------
 Returned values
 -------------------------------------------------------------------------
 TuningOptions(1) - Number of decimals allowed
              (2) - Step coefficient (initial step coeff.)
              (3) - Similarity treshold
              (4) - allowed iteration number
              (5) - Info display during iteration
              (6) - Save FIS and info after each calculation step
              (7) - IncrementTreshold
                    Increase (double) step if MSE changes more then 
                    IncrementTreshold during an iteration
              (8) - DecrementTreshold
                    Decrease (halve) step if MSE does not change more than 
                    DecrementTreshold during an iteration
              (9) - Merge simlar sets after each iteration
             (10) - Merge similar rules after each iteration
             (11) - Plot RMSE (default: 1)
             (12) - Plot original and calculated data
             (13) - Tuning method type
                     1 - RBE-DSS
                     2 - RBE-SI
                     3 - ACP
                     4 - Simplex
             (14) - Performance index type
                     1 - MSE
                     2 - RMSE
                     3 - RMSEP
                     4 - R
                     5 - RR
                     6 - AD
                     7 - ADP
                     8 - SKL Simmetrized Kulback Leibler divergence
                     9 - JDD Jensen Difference Divergence measure
             (15) - Stop the iteration when the performance index falls below this value
             (16) - Core width in percentage of the range
             (17) - Support width in percentage of the range  
             (18) - Parameterization type
                     1 - breakpoints
                     2 - relative distances
                     3 - conserving Ruspini partition
                     4 - reference point
             (19) - Minimum step (by the simplex method) expressed in
                    proportion of the range of the current linguistic 
                    variable.
 -------------------------------------------------------------------------
 Remarks
 -------------------------------------------------------------------------

 Zsolt Csaba Johanyák, johanyak.csaba@gamf.kefo.hu, v. 1.5,  22 Oct 2008

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:
Generated on Thu 26-Mar-2009 13:11:50 by m2html © 2005