From 08fd3f24a7cda71a7a9c6a38ef2e0481c525d211 Mon Sep 17 00:00:00 2001
From: Sepehr Madani <ssepehrmadani@gmail.com>
Date: Mon, 27 Jul 2020 18:51:49 -0400
Subject: [PATCH] Merge GA and GB into a single class
---
algorithms/genetic_algorithm.py | 171 ++++++++++++++++++++++++--------
1 file changed, 129 insertions(+), 42 deletions(-)
diff --git a/algorithms/genetic_algorithm.py b/algorithms/genetic_algorithm.py
index 87bc1e0..aeeafab 100644
--- a/algorithms/genetic_algorithm.py
+++ b/algorithms/genetic_algorithm.py
@@ -1,7 +1,8 @@
-import cmath
-import random
+from random import randrange, random, choice, sample
+from cmath import exp, phase
from copy import deepcopy
-from math import cos, degrees, inf, log10, pi, radians, sin
+from math import log10, pi
+from typing import List
from utils.pattern import compute_pattern
@@ -9,13 +10,18 @@ from .base_algorithm import BaseAlgorithm
class Chromosome:
- def __init__(self, N, bit_count):
- self.gene = [Chromosome.new_gene(bit_count) for i in range(N)]
+ def __init__(self, n, bit_count):
+ self.gene = [Chromosome.new_gene(bit_count) for _ in range(n)]
self.fitness = float("nan")
+ self.needs_update = True
+
+ def get_score(self):
+ """Evaluates a score based on chromosome's fitness"""
+ return -20 * log10(abs(self.fitness))
@staticmethod
def new_gene(bit_count):
- return random.randrange(0, 2 ** bit_count)
+ return randrange(0, 2 ** bit_count)
class GeneticAlgorithm(BaseAlgorithm):
@@ -23,8 +29,10 @@ class GeneticAlgorithm(BaseAlgorithm):
discrete values.
"""
+ chromosomes: List[Chromosome]
+
def __init__(self, options):
- BaseAlgorithm.__init__(self, options)
+ super().__init__(options)
self.main_ang = options.main_ang
self.sample_size = options.sample_size
self.null_degrees = options.null_degrees
@@ -32,67 +40,146 @@ class GeneticAlgorithm(BaseAlgorithm):
self.bit_count = options.bit_count
self.bit_resolution = options.bit_resolution
self.mutation_factor = options.mutation_factor
+ self.overwrite_mutations = options.overwrite_mutations
+
+ self.chromosomes = []
+
+ self.buckets = None
+ if options.use_buckets:
+ self.bucket_count = options.bucket_count
+ self.buckets = [[]] * self.bucket_count
self.check_parameters()
def check_parameters(self):
super().check_parameters()
+ if self.buckets is not None:
+ assert len(self.null_degrees) == 1
+ assert self.bucket_count & 1 == 0
def solve(self):
- self.intialize_sample()
- self.update_fitness()
- self.sort_fitness()
+ self.initialize_sample()
+ self.organize_sample()
+
for generation in range(self.gen_to_repeat):
- for ii in range(self.sample_size // 2, self.sample_size - 1, 2):
- p1, p2 = random.sample(range(self.sample_size // 2), 2)
- self.crossover(p1, p2, ii, ii + 1)
+ self.create_children()
self.mutate_sample()
- self.update_fitness()
- self.sort_fitness()
-
- # print(["{:.2f}".format(x.fitness) for x in self.chromosomes[:15]])
- return self.make_weights(self.chromosomes[0])
-
- def mutate_sample(self):
- for chromosome in self.chromosomes[1:]: # for all except the best chromosome
- for idx in range(self.N):
- if random.random() <= self.mutation_factor:
- chromosome.gene[idx] = Chromosome.new_gene(self.bit_count)
+ self.organize_sample()
+
+ return (self.make_weights(self.chromosomes[0]), self.chromosomes[0].get_score())
+
+ def create_children(self):
+ """Using the better half of the population, creates children overwriting the bottom half by doing crossovers.
+ If use_buckets is True, uses AM-GM–based crossover. Otherwise, it uses the basic merger crossover."""
+
+ for child in range(self.sample_size // 2, self.sample_size - 1, 2):
+ if self.buckets is None:
+ p1, p2 = sample(range(self.sample_size // 2), 2)
+ self.crossover(p1, p2, child, child + 1)
+ else:
+ bucket_idx = randrange(self.bucket_count)
+ p1 = choice(self.buckets[bucket_idx])
+ p2 = min(
+ self.buckets[(bucket_idx + self.bucket_count//2) % self.bucket_count],
+ key=lambda x: abs(x.fitness + p1.fitness)
+ )
+ self.crossover_bucket(p1, p2, child, child + 1)
- def update_fitness(self, use_exact_angle=True):
+ def organize_sample(self):
+ """Reorganizes the sample by updating fitness for all chromosomes and sorting them by their scores.
+ Optionally, if use_buckets is True, allocates each chromosome to its respective bucket."""
+
+ # Update fitness
for chromosome in self.chromosomes:
- values = [
- -20 * log10(abs(x))
- for x in compute_pattern(
- N=self.N,
- k=self.k,
- weights=self.make_weights(chromosome),
- degrees=self.null_degrees,
+ if chromosome.needs_update:
+ chromosome.fitness = min(
+ compute_pattern(
+ N=self.N,
+ k=self.k,
+ weights=self.make_weights(chromosome),
+ degrees=self.null_degrees,
+ use_absolute_value=False
+ )
)
- ]
- chromosome.fitness = min(values)
+ chromosome.needs_update = False
+
+ # Sort sample by fitness
+ self.chromosomes.sort(key=lambda x: x.get_score(), reverse=True)
- def sort_fitness(self):
- self.chromosomes.sort(key=lambda x: x.fitness, reverse=True)
+ # Allocate chromosomes to their respective buckets
+ if self.buckets is not None:
+ self.buckets.clear()
+ self.buckets = [[]] * self.bucket_count
+ for chromosome in self.chromosomes:
+ bucket_idx = int(((phase(chromosome.fitness) + pi) / (2 * pi)) * self.bucket_count) % self.bucket_count
+ self.buckets[bucket_idx].append(chromosome)
+
+ def mutate_sample(self):
+ """Mutates the sample excluding the best chromosome.
+ Overwrites the previous chromosomes if overwrite_mutations is True."""
+
+ if self.overwrite_mutations:
+ # For all except the best chromosome
+ for chromosome in self.chromosomes[1:]:
+ chromosome.needs_update = True
+ for idx in range(self.N):
+ if random() <= self.mutation_factor:
+ chromosome.gene[idx] = Chromosome.new_gene(self.bit_count)
+ else:
+ # For all except the best chromosome
+ for original in range(1, self.sample_size):
+ mutated = original + self.sample_size - 1
+ self.chromosomes[mutated].needs_update = True
+
+ for ii in range(self.N):
+ if random() <= self.mutation_factor:
+ self.chromosomes[mutated].gene[ii] = Chromosome.new_gene(self.bit_count)
+ else:
+ self.chromosomes[mutated].gene[ii] = self.chromosomes[original].gene[ii]
def make_weights(self, chromosome):
+ """Returns e^{iθ} value for a chromosome's θs"""
+
weights = []
for bits in chromosome.gene:
angle = (bits - (2 ** self.bit_count - 1) / 2) * 2 * pi / (2 ** self.bit_resolution)
- weights.append(cmath.exp(1j * angle))
+ weights.append(exp(1j * angle))
return weights
def crossover(self, p1, p2, c1, c2):
+ """Merges two parents' genes to create two children"""
+
self.chromosomes[c1] = deepcopy(self.chromosomes[p1])
self.chromosomes[c2] = deepcopy(self.chromosomes[p2])
for i in range(self.N):
- if random.random() >= 0.5:
+ if random() >= 0.5:
self.chromosomes[c1].gene[i], self.chromosomes[c2].gene[i] = (
self.chromosomes[c1].gene[i],
self.chromosomes[c2].gene[i],
)
- def intialize_sample(self):
- self.chromosomes = [
- Chromosome(self.N, self.bit_count) for i in range(self.sample_size)
- ]
\ No newline at end of file
+ def crossover_bucket(self, p1, p2, c1, c2):
+ """Creates two children from parents' genes using AM-GM"""
+
+ for ii in range(self.N):
+ g1 = p1.gene[ii]
+ g2 = p2.gene[ii]
+ self.chromosomes[c1].gene[ii] = (g1 + g2) // 2
+ self.chromosomes[c2].gene[ii] = (g1 + g2 + 1) // 2
+
+ def initialize_sample(self):
+ """Destroys all chromosomes and creates a new random population"""
+
+ self.chromosomes.clear()
+ if self.overwrite_mutations:
+ self.chromosomes = [
+ Chromosome(self.N, self.bit_count) for _ in range(self.sample_size)
+ ]
+ else:
+ self.chromosomes = [
+ Chromosome(self.N, self.bit_count) for _ in range(self.sample_size * 2 - 1)
+ ]
+
+ if self.buckets is not None:
+ self.buckets.clear()
+ self.buckets = [[]] * self.bucket_count
--
GitLab