@micgeronimo All contributions are welcomed. I think regression tests are what is most pressing. For example, benchmarking the convergence of some standard examples could help spot if there is a breaking change pretty easily.
Hi, I am running some tests to checkout DEAP. I have this code to instantiate a simple population of simple individuals. The individuals have features that is 1 list of integers. Fitness is calculated by summing the count of feature integers that match a goal integer. And then I run selStochasticUniversalSampling to return a new list.
import numpy as np
import deap.tools as tools
class Individual():
def __init__(self):
self.features = np.random.randint(low=0,high=10,size=5)
self.goal = 5
self.fitness = 0
def fitness_calc(self):
self.fitness = sum(self.features==self.goal)
class Population():
def __init__(self):
self.count_individual = 100
self.list_individual = []
for i in range(self.count_individual):
self.list_individual.append(Individual())
def fitness_calc(self):
for individual in self.list_individual:
individual.fitness_calc()
my_population = Population()
my_population.fitness_calc()
sel_population = tools.selStochasticUniversalSampling(individuals=my_population.list_individual,
k=10,
fit_attr="fitness")
print("complete")This fails with traceback:
Traceback (most recent call last):
File "/home/calvin/PycharmProjects/go/go_test.py", line 28, in <module>
sel_population = tools.selStochasticUniversalSampling(individuals=my_population.list_individual,k=10,fit_attr="fitness")
File "/home/calvin/.local/lib/python3.6/site-packages/deap/tools/selection.py", line 197, in selStochasticUniversalSampling
sum_fits = sum(getattr(ind, fit_attr).values[0] for ind in individuals)
File "/home/calvin/.local/lib/python3.6/site-packages/deap/tools/selection.py", line 197, in <genexpr>
sum_fits = sum(getattr(ind, fit_attr).values[0] for ind in individuals)
AttributeError: 'numpy.int64' object has no attribute 'values'Does the fit_attr have to be of a certain datatype? Is there a DEAP fitness datatype that must be used?
I need 2 floats, 1 bool and 1 int
@NirvanaDogra yes, plenty of people use GA to analyse data for the financial market
@jcbro not the best place to post code try the mailing list with gist or Pastebin
Hello everyone :) !
I am working on a problem that requires creating a function with n inputs and k outputs. From what I know a GP tree can have only one output, so I thought I would use a list of k trees each getting the same n inputs and producing one of the k outputs.
It seems like a good idea but I am hesitant if it is a correct approach. How will crossover and mutation operators work in case of a list of trees? Will that make sense? Maybe there is a better way?
Source code I have so far below:
I am working on a problem that requires creating a function with n inputs and k outputs. From what I know a GP tree can have only one output, so I thought I would use a list of k trees each getting the same n inputs and producing one of the k outputs.
It seems like a good idea but I am hesitant if it is a correct approach. How will crossover and mutation operators work in case of a list of trees? Will that make sense? Maybe there is a better way?
Source code I have so far below:
pset = gp.PrimitiveSet("MAIN", arity=n, prefix='x')
pset.addPrimitive(operators.addition, 2)
pset.addPrimitive(operators.subtract, 2)
pset.addPrimitive(operators.multiply, 2)
# and many more...
# One optimization objective.
creator.create("FitnessMin", base.Fitness, weights=(-1.0,))
# Arithmetic expression.
creator.create("Tree", gp.PrimitiveTree, pset=pset)
# Individual is a list consisting of n_components trees.
creator.create("Individual", list, fitness=creator.FitnessMin)
toolbox = base.Toolbox()
toolbox.register("expr", gp.genHalfAndHalf, pset=pset, min_=MIN_TREE_HEIGHT, max_=MAX_TREE_HEIGHT)
toolbox.register("tree", tools.initIterate, creator.Tree, toolbox.expr)
toolbox.register("individual", tools.initRepeat, creator.Individual, toolbox.tree, n=pop_size)
# Population.
toolbox.register("population", tools.initRepeat, list, toolbox.individual)
Hi, I am working with genetic algorithm in DEAP. I want to parallelize the evalution part that's why I am using multiprocessing. The problem is evaluation function works for only one individual and after that it throws an error above ( Can't pickle <function evaluation at 0x000001ED70DA28C8>: it's not the same object as ga_evaluation.evaluation). How can I solve this issue, any ideas?
@buseoz_gitlab That is a function pickling problem. Functions are tricky to pickle, but from you error message here is what I would try first : https://stackoverflow.com/questions/1412787/picklingerror-cant-pickle-class-decimal-decimal-its-not-the-same-object
Hello, I'm new with EAP, but I have got a very complex individual (as a class), My class has got a mutate and combine method and a method which return two fitness values which should be maximized. In my case the constructure of the class need some initializing parameters. I don't find a tutorial which explain to use the Toolbox with a individual class. Can anybody help me?
I need only the "population creation" and the "individual selection" calls from the toolbox, I would like to use the SwarmParticle System https://deap.readthedocs.io/en/master/examples/pso_basic.html because I have got 2 maximazing parameters
In my case I need an own generator which returns a list of my class individuals, an own mutate & combine function and an own evaluate function
How can I do this?
@FranciscoPalomares use the feasibility function as in the constraint tutorial.
def feasible(individual):
"""Feasability function for the individual. Returns True if feasible False
otherwise."""
if int(individual[2]) < int(individual[1]):
return True
return False
toolbox.register("evaluate", evalOneMax)
toolbox.decorate("evaluate", tools.DeltaPenality(feasible,7.0))@FranciscoPalomares Documentation for DeltaPenalty and tutorial . 7.0 is the penalty applied to invalid individuals. It is a required argument.
Diversity strongly depends on your response domain. If your population converges too fast you can try to play with crossover and mutation parameters. Moreover having selection with randomness (like tournament with a low number of participant) may help.
Hello, thanks for this wonderful tool!
I have two parameters that I want to maximize with some fitness function. But each parameter has a certain limit to its range, i.e. parameter value is limit between -1 to 100, and parameter b range limit is between -50 to -10.
How can I limit the evolution of values to a certain range?
Thanks to the helpers :-)
I have two parameters that I want to maximize with some fitness function. But each parameter has a certain limit to its range, i.e. parameter value is limit between -1 to 100, and parameter b range limit is between -50 to -10.
How can I limit the evolution of values to a certain range?
Thanks to the helpers :-)
@Hananel-Hazan Hey, you can either use soft constraints, operators that handle bounds (bounded operators), or repair in your evaluation function.
@nbro I suggest you try to run the first few lines of the symbreg and generate individuals one at a time to see what they look like.
You'll get a sense of how it works. From there you can move to ant (which is a lot more complex)