| The Numerical Constants Tab gives access to the
following settings and features:
Use Random Numerical Constants
Check this box if you want to use the GEP-RNC learning algorithm of
APS. This algorithm has a facility for handling random numerical constants and
it evolves models with them.
Constants per Gene
The number of random numerical constants associated with each
gene.
Data Type
The data type (integer or floating-point) you think appropriate
for the random numerical constants that will be used in the creation
of your models. For Integer constants only integer values in the
interval [-2147483648, 2147483647] are acceptable. For
Floating-point constants a double-precision floating-point value
with a range of -1.79769313486231E308 to -4.94065645841247E-324 for
negative values, 4.94065645841247E-324 to 1.79769313486231E308 for
positive values, and 0 are acceptable.
Lower Bound
Sets the lower bound of the interval from which random numerical
constants are drawn.
Upper Bound
Sets the upper bound of the interval from which random numerical
constants are drawn.
RNC Mutation
The RNC mutation rate used to create genetic variation in the arrays of random numerical constants. The default is 0.01.
Dc Mutation
The Dc mutation rate used to create genetic variation. The default is 0.044.
Dc Inversion
The Dc inversion rate used to create genetic variation. The default is 0.1.
Dc IS Transposition
The Dc IS transposition rate used to create genetic variation. The default is 0.1.
Defaults
Resets the rates of all the above genetic operators to the default values.
|