Global Index (all files) (short | long) | Local contents | Local Index (files in subdir) (short | long)
fitdistc(OBJ_F, VLUB, FitDistOpt, Dummy, P1,P2,P3,P4,P5,P6,P7,P8,P9,P10,P11,P12,P13,P14,P15)
FITness DISTance Correlation computation This function computes the fitness-distance correlation of a given objective function or using saved data. A fitness distance scatter plot is generated. Syntax: fitdistc(OBJ_F, VLUB, FitDistOpt, Dummy, P1,P2,P3,P4,P5,P6,P7,P8,P9,P10,P11,P12,P13,P14,P15) Input parameters: OBJ_F - String containing the name of the objective function VLUB - (optional) Matrix containing boundaries of variables of objective function if VLUB is empty, feval(OBJ_F, [],1) is used to determine the boundaries to use if OBJ_F is empty, VLUB contains the full name of a text data file containing objective values (first column) and variables of corresponding individuals (second - last column), each row corresponds to objective value and variables of one individual FitDistOpt- (optional) Vektor containing options of fitness distance calculation FitDistOpt(1): Number of variables (dimensions) to use if omitted, number of variables is determined from size of VLUB FitDistOpt(2): 0: Use points generated by an evenly spaced grid 1: Use randomly generated points if number of variables (dimensions) is greater than 6, randomly created points are used FitDistOpt(3): Maximal number of points to use Dummy - not used at the moment P1-P15 - Additional parameters directly forwarded to the objective function Output parameter: no output parameter Examples: % Compute fdc of objfun1 using default parameters >> fitdistc('objfun1') % Compute fdc of objfun1 in 5 dimensions, use 10000 evenly spaced points >> fitdistc('objfun1', [], [5, 0, 10000]) See also: geamain, resplot, compdiv
This function calls | |
---|---|