[Tutor] Tutor Digest, Vol 35, Issue 38

[Carlos]

Hello,

> Unfortunately, no. I for one can't interpret "I need the info every 
> time that it gets updated by the loop"
The GA is creating solutions for a given problem, that are contained in 
self.entities that looks like this:

Generation_01: [[1, 0, 0, 1, 0, 1, 0, 0, 0, 0], [0, 0, 1, 1, 1, 1, 1, 0, 
0, 0], [1, 1, 0, 0, 1, 1, 1, 1, 1, 1], [0, 1, 1, 1, 1, 0, 0, 1, 0, 1], 
[1, 0, 1, 0, 1, 0, 1, 1, 0, 1]]
Generation_02, Generation_03, Generation_04...
Generation_25:[[1, 1, 1, 1, 1, 1, 1, 0, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 
1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1], 
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]]

self.entities gets new values with each generation, because the solution 
to a given problem is evolving and getting closer to a correct result. 
The problem is that with:

 Phenotypes = ga_inst.evolve(Gen)

I only get the last generation -or iteration- and I need to actualize 
some objects position before evaluating the generation, so what I need 
is to get the values of self.entity on each generation. Since this 
values are going to be evaluted, I need to actualize some gemetrical 
objects with every new self.entities before evaluating the solution.

I tried this:

def evolve(self, generations = 100):
        """ Process a number of generations
        Stop if:
        - number of generations = generations
        - halt variable has been set AND
          there is an entity with a fitness which has obtained or exceeded
          the halt value
        Return the list of entities, or the halt-entity if prematurely 
halted
          """
       
        self.halt_reached = 0
       
        for gen in range(generations):

            print 'TEST', self.entities ### THIS IS MY CHANGE


            self.do_mutation()
            self.do_crossover()
            print "Average fitness generation " + str(gen) + ": " + 
str(self.avg_fitness())
            if self.debug == 1:
                self.echo_fitness()
            if self.halt >= 0:
                max_entity = self.get_max_fitness()
                fit = self.calc_fitness(max_entity)
                if fit >= halt:
                    self.halt_reached = 1
                    return [max_entity]
               
        return self.entities

I can actually print the self.entities list once for each iteration, 
before the evaluation takes place. What I dont know is how to actually 
do something with the list, like putting it in a variable or something 
to get access to it and actualize the geometry of my project. So 
basically if there are 100 iterations I would need a way to get this 
list 100 times, one for each iteration with its new data.

And just in case that it helps, the address of the forum were I uploaded 
a piece of the code:

http://python-forum.org/py/viewtopic.php?p=11063#11063

And the GA project page:

Genetic Algorithm Python module
http://www.alextreme.org/projects/python_ai/

If you can help me please do it, this is critical for my project :(

Thanks for helping
Carlos