def runGA(genomeFilepath, resultsFilepath, numSpecies, speciesSize, mutationRate, evalTime, numEnvs, numGens):
allFitnessScores = {}
envs = ENVIRONMENTS(numEnvs)
parents = initialize_population(numSpecies, speciesSize, mutationRate, evalTime, numEnvs)
for g in range(1, numGens+1):
children = SWARM(numSpecies, speciesSize, mutationRate, evalTime, numEnvs)
children.Fill_From(parents)
children.evaluateSwarms(envs, pp=False, pb=True)
print(g),
children.Print()
parents = children
allFitnessScores[g] = parents.getFitnesses()
pickle.dump(allFitnessScores, open(os.path.join("results", resultsFilepath + ".txt"), "wb"))
pickle.dump(parents, open(os.path.join("results", genomeFilepath + ".txt"), "wb"))
def main():
envs = ENVIRONMENTS(c.numEnvs)
if (c.loadPickledPopulation and os.path.isfile("save.txt")):
parents = pickle.load(open("save.txt", "rb"))
else:
parents = initialize_population()
for g in range(1, c.numGens+1):
children = SWARM(c.numSpecies, c.speciesSize, c.mutationRate, c.evalTime, c.numEnvs)
children.Fill_From(parents)
children.evaluateSwarms(envs, pp=False, pb=True)
print(g),
children.Print()
parents = children
pickle.dump(parents, open("save.txt", "wb"))
from __future__ import print_function
import pyrosim
import matplotlib.pyplot as plt
import random
import copy
from robot import ROBOT
import pickle
from population import POPULATION
from individual import INDIVIDUAL
import constants as c
from environments import ENVIRONMENTS
from environment import ENVIRONMENT
envs = ENVIRONMENTS()
parents = POPULATION(c.popSize)
parents.Initialize()
parents.Evaluate(envs, pp=False, pb=True)
#
parents.Print()
for g in range(1, c.numGens):
children = POPULATION(c.popSize)
children.Fill_From(parents)
children.Evaluate(envs, pp=False, pb=True)
print(g),
children.Print(),
print(),
parents.ReplaceWith(children)
import random
import copy
import logging
logging.basicConfig(level=logging.INFO)
import pickle
import pyrosim
import matplotlib.pyplot as plt
import numpy as np
from robot import ROBOT
from individual import INDIVIDUAL
from population import POPULATION
import constants
from environments import ENVIRONMENTS
# create environment with "light source"
envs = ENVIRONMENTS(numEnvs=constants.numEnvs, eval_time=constants.eval_time)
# create initial population
parents = POPULATION(envs,
popSize=constants.popSize,
eval_time=constants.eval_time)
parents.initialize()
parents.evaluate(play_blind=True, play_paused=False)
print(0, parents)
# evolve:
for g in range(constants.numGen):
children = POPULATION(envs, parents=parents)
children.fillFrom(parents)
children.evaluate(play_blind=True, play_paused=False)
parents.replaceWith(children)
from environments import ENVIRONMENTS
from swarm import SWARM
import constants as c
import pickle
envs = ENVIRONMENTS(c.numEnvs)
parents = pickle.load(open("save.txt", "rb"))
parents.evaluateSwarms(envs, pp=True, pb=False)
def initialize_population(numSpecies, speciesSize, mutationRate, evalTime, numEnvs): parents = SWARM(numSpecies, speciesSize, mutationRate, evalTime, numEnvs) parents.Initialize() envs = ENVIRONMENTS(numEnvs) parents.evaluateSwarms(envs, pp=True, pb=True) return (parents)
if play_result:
parents.evaluate(envs, best=True)
if store_fitness is not None:
parents.p[0].print(to=store_fitness)
def test_loop(pop_range: list, t_fraction: float, m_fraction: float):
with open('data_out3.csv', 'w') as output:
output.write('pop_size,t_size,m_rate,score')
for pop_size in range(*pop_range):
for t_size in range(1,
int(t_fraction * pop_size) + 1,
int(pop_size / 10)):
C.set_cons('t_size', t_size)
for m_rate in range(1, int(m_fraction * 72) + 1, int(72 / 10)):
C.set_cons('m_rate', m_rate)
output.write(f'\n{pop_size},{t_size},{m_rate}')
genetic_algorithm(
pop_size, f'playbacks2/p{pop_size}t{t_size}m{m_rate}',
False, output)
envs = ENVIRONMENTS() if C.numEnvs > 0 else None
# test_loop([25, 26, 5], 0.7, 0.5)
genetic_algorithm()