/
ga_runner.py
139 lines (116 loc) · 5.51 KB
/
ga_runner.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
import genetic_algorithm
import corners
import utils
import sys
import random
import time
import multiprocessing
import numpy as np
if __name__ == "__main__" and len(sys.argv) == 10:
cornerFile = sys.argv[1]
cornerSquares = corners.readCornersFromFile(cornerFile)
totalN = int(sys.argv[2])
n = totalN - sum([len(c) for c in cornerSquares])
populationSize = int(sys.argv[3])
matingPoolSize = int(sys.argv[4])
numGenerations = int(sys.argv[5])
mutationProbability = float(sys.argv[6])
mutationPerBitProbability = float(sys.argv[7])
crossoverProbability = float(sys.argv[8])
selectionPressure = float(sys.argv[9])
outputFilename = "garuns/{n}/{populationSize}_{matingPoolSize}_{numGenerations}_{mutationProbability}_{mutationPerBitProbability}_{crossoverProbability}_{selectionPressure}".format(
n=totalN,
populationSize=populationSize,
matingPoolSize=matingPoolSize,
numGenerations=numGenerations,
mutationProbability=mutationProbability,
mutationPerBitProbability=mutationPerBitProbability,
crossoverProbability=crossoverProbability,
selectionPressure=selectionPressure
)
population = [genetic_algorithm.generateRandomChromosome(n) for _ in range(populationSize)]
generation = 0
# logs = []
# tempLogs = {}
generationSummary = []
matingTime = 0
while generation <= numGenerations:
print("Running generation", generation + 1, "of", numGenerations)
startTime = time.time()
# fitnesses = [0] * populationSize
# threads = []
# for i in range(populationSize):
# th = multiprocessing.Process(target=genetic_algorithm.parallel_fitness_helper, args=(i, cornerSquares, population[i], fitnesses, logs, generation != 0))
# threads.append(th)
# th.start()
# for th in threads:
# th.join()
# for i in range(populationSize):
# genetic_algorithm.parallel_fitness_helper(i, cornerSquares, population[i], fitnesses, tempLogs, generation != 0)
fitnesses = [genetic_algorithm.fitness_function(cornerSquares, chromosome) for chromosome in population]
fitnessTime = time.time() - startTime
# for k, v in tempLogs.items():
# logs.append(v)
fitnesses, population = zip(*sorted(zip(fitnesses, population)))
currentSummary = {
"topFitnesses": fitnesses[:5],
"topChromosomes": population[:5],
"generation": generation,
"bestFitness": min(fitnesses),
"10percentileFitness": np.percentile(fitnesses, 10),
"25percentileFitness": np.percentile(fitnesses, 25),
"medianFitness": np.median(fitnesses),
"fitnessTime": fitnessTime,
"matingTime": matingTime
}
generationSummary.append(currentSummary)
if generation == numGenerations:
# so we have analysis of last generation
break
startTime = time.time()
matingPool = genetic_algorithm.rouletteWheelSelection(population, fitnesses, matingPoolSize, selectionPressure)
nextGeneration = []
while len(nextGeneration) < populationSize:
randIdx1 = random.randint(0, matingPoolSize - 1)
randIdx2 = random.randint(0, matingPoolSize - 1)
parent1 = matingPool[randIdx1]
parent2 = matingPool[randIdx2]
avgParentFitness = (fitnesses[randIdx1] + fitnesses[randIdx2]) / 2
crossed = False
if random.random() < crossoverProbability:
crossed = True
crossoverPoint = random.randint(0, len(parent1) - 1)
newChromosome1, newChromosome2 = genetic_algorithm.crossover(parent1, parent2, crossoverPoint, len(parent1))
else:
newChromosome1 = parent1
newChromosome2 = parent2
mutated1 = False
mutated2 = False
if random.random() < mutationProbability:
mutated1 = True
newChromosome1, mutations1 = genetic_algorithm.mutationFlipbits(newChromosome1, mutationPerBitProbability)
if random.random() < mutationProbability:
mutated2 = True
newChromosome2, mutations2 = genetic_algorithm.mutationFlipbits(newChromosome2, mutationPerBitProbability)
nextGeneration.append(newChromosome1)
nextGeneration.append(newChromosome2)
# utils.saveChromosomeInJson(tempLogs,
# newChromosome1,
# generation + 1,
# (parent1, parent2) if crossed else parent1,
# avgParentFitness if crossed else fitnesses[randIdx1],
# mutations1 if mutated1 else [],
# (crossoverPoint, len(parent1)) if crossed else (-1, -1))
# utils.saveChromosomeInJson(tempLogs,
# newChromosome2,
# generation + 1,
# (parent1, parent2) if crossed else parent2,
# avgParentFitness if crossed else fitnesses[randIdx2],
# mutations2 if mutated2 else [],
# (crossoverPoint, len(parent2)) if crossed else (-1, -1))
matingTime = time.time() - startTime
print("Generation", generation + 1, "done")
population = nextGeneration
generation += 1
# utils.saveLogsToFile(logs, outputFilename + "_chromosomes.json")
utils.saveLogsToFile(generationSummary, outputFilename + "_summary.json")