There are 2 ways to implement continuous chromosomes in GALIB32.DLL:
1. Create each chromosome individually using the MakeChromosome function. In this case, you must call the function each time you want to create a continuous chromosome.
2. Create a chromosome pool using the MakeChromosomePool function. This function is very useful in cases where you need to create a large number of chromosomes at one time. If you call the function once, it will create a large number of completely identical continuous chromosomes.
A chromosome pool is an array of identical continuous chromosomes. Each parameter (independent variable) of the problem corresponds to a separate chromosome in the chromosome pool. The chromosome pool is useful especially in those cases when a very large number (1,000-10,000) of identical continuous chromosomes must be created. Such a situation will occur, for example, when GeneHunter is applied to the training of a neural net. If you’re using a GA to select the weights for a neural network, the use of chromosome pools will be more efficient. Each weight of the net is considered to be a continuous chromosome in the pool.
1. Integers within a chromosome pool created by the MakeChromosomePool function are not available.
2. An advanced operator, Diversity, cannot be applied to the chromosome pool during evolution. The Diversity operator may be applied only to those individual chromosomes created by the MakeChromosome function.