Deap: python中的遗传算法工具箱

Overview 程序概览

官方文档:http://deap.readthedocs.io/en/master/index.html

1. Types : 选择你要解决的问题类型,确定要求解的问题个数,最大值还是最小值
2. Initialization : 初始化基因编码位数,初始值,等基本信息
3. Operators : 操作,设计evaluate函数,在工具箱中注册参数信息:交叉,变异,保留个体,评价函数
4. Algorithm : 设计main函数,确定参数并运行得到结果

Types

# Types
from deap import base, creator

creator.create("FitnessMin", base.Fitness, weights=(-1.0,))  

# weights 1.0, 求最大值,-1.0 求最小值
# (1.0,-1.0,)求第一个参数的最大值,求第二个参数的最小值
creator.create("Individual", list, fitness=creator.FitnessMin)

Initialization

import random
from deap import tools

IND_SIZE = 10  # 种群数

toolbox = base.Toolbox()
toolbox.register("attribute", random.random)
# 调用randon.random为每一个基因编码编码创建 随机初始值 也就是范围[0,1]
toolbox.register("individual", tools.initRepeat, creator.Individual,
                 toolbox.attribute, n=IND_SIZE)
toolbox.register("population", tools.initRepeat, list, toolbox.individual)

Operators

# Operators
# difine evaluate function
# Note that a comma is a must
def evaluate(individual):
    return sum(individual),


# use tools in deap to creat our application
toolbox.register("mate", tools.cxTwoPoint) # mate:交叉
toolbox.register("mutate", tools.mutGaussian, mu=0, sigma=1, indpb=0.1) # mutate : 变异
toolbox.register("select", tools.selTournament, tournsize=3) # select : 选择保留的最佳个体
toolbox.register("evaluate", evaluate)  # commit our evaluate

高斯变异:

这种变异的方法就是，产生一个服从高斯分布的随机数，取代原先基因中的实数数值。这个算法产生的随机数，数学期望当为当前基因的实数数值。
一个模拟产生的算法是，产生6个服从U(0,1)的随机数，以他们的数学期望作为高斯分布随机数的近似。

mutate方法

• 这个函数适用于输入个体的平均值和标准差的高斯突变

• mu:python中基于平均值的高斯变异

• sigma:python中基于标准差的高斯变异

• indpb:每个属性的独立变异概率

mate : 交叉

select : 选择保留的最佳个体

evaluate : 选择评价函数,要注意返回值的地方最后面要多加一个逗号

Algorithms 计算程序

也就是设计主程序的地方,按照官网给的模式,我们要早此处设计其他参数,并设计迭代和取值的代码部分,并返回我们所需要的值.

# Algorithms
def main():
    # create an initial population of 300 individuals (where
    # each individual is a list of integers)
    pop = toolbox.population(n=50)
    CXPB, MUTPB, NGEN = 0.5, 0.2, 40

    '''
    # CXPB  is the probability with which two individuals
    #       are crossed
    #
    # MUTPB is the probability for mutating an individual
    #
    # NGEN  is the number of generations for which the
    #       evolution runs
    '''

    # Evaluate the entire population
    fitnesses = map(toolbox.evaluate, pop)
    for ind, fit in zip(pop, fitnesses):
        ind.fitness.values = fit

    print("  Evaluated %i individuals" % len(pop))  # 这时候，pop的长度还是300呢
    print("-- Iterative %i times --" % NGEN)

    for g in range(NGEN):
        if g % 10 == 0:
            print("-- Generation %i --" % g)
        # Select the next generation individuals
        offspring = toolbox.select(pop, len(pop))
        # Clone the selected individuals
        offspring = list(map(toolbox.clone, offspring))
        # Change map to list,The documentation on the official website is wrong

        # Apply crossover and mutation on the offspring
        for child1, child2 in zip(offspring[::2], offspring[1::2]):
            if random.random() < CXPB:
                toolbox.mate(child1, child2)
                del child1.fitness.values
                del child2.fitness.values

        for mutant in offspring:
            if random.random() < MUTPB:
                toolbox.mutate(mutant)
                del mutant.fitness.values

        # Evaluate the individuals with an invalid fitness
        invalid_ind = [ind for ind in offspring if not ind.fitness.valid]
        fitnesses = map(toolbox.evaluate, invalid_ind)
        for ind, fit in zip(invalid_ind, fitnesses):
            ind.fitness.values = fit

        # The population is entirely replaced by the offspring
        pop[:] = offspring

    print("-- End of (successful) evolution --")

    best_ind = tools.selBest(pop, 1)[0]

    return best_ind, best_ind.fitness.values  # return the result:Last individual,The Return of Evaluate function
    

要注意的地方就是,官网中给出的Overview代码中有一行代码是错误的,需要把一个数据类型(map)转换为list.

输出结果

  Evaluated 50 individuals
-- Iterative 40 times --
-- Generation 0 --
-- Generation 10 --
-- Generation 20 --
-- Generation 30 --
-- End of (successful) evolution --
best_ind [-2.402824207878805, -1.5920248739487302, -4.397332290574777, -0.7564815676249151, -3.3478264358788814, -5.900475519316307, -7.739284213710048, -4.469259215914226, 0.35793917907272843, -2.8594709616875256]
best_ind.fitness.values (-33.10704010746149,)

• best_ind : 最佳个体
• best_ind.fitness.values : 最佳个体在经过evaluate之后的输出

#!usr/bin/env python
# -*- coding:utf-8 _*-
"""
@author:fonttian 
@file: Overview.py
@time: 2017/10/15 
"""

# Types
from deap import base, creator

creator.create("FitnessMin", base.Fitness, weights=(-1.0,))
# weights 1.0, 求最大值,-1.0 求最小值
# (1.0,-1.0,)求第一个参数的最大值,求第二个参数的最小值
creator.create("Individual", list, fitness=creator.FitnessMin)

# Initialization
import random
from deap import tools

IND_SIZE = 10  # 种群数

toolbox = base.Toolbox()
toolbox.register("attribute", random.random)
# 调用randon.random为每一个基因编码编码创建 随机初始值 也就是范围[0,1]
toolbox.register("individual", tools.initRepeat, creator.Individual,
                 toolbox.attribute, n=IND_SIZE)
toolbox.register("population", tools.initRepeat, list, toolbox.individual)


# Operators
# difine evaluate function
# Note that a comma is a must
def evaluate(individual):
    return sum(individual),


# use tools in deap to creat our application
toolbox.register("mate", tools.cxTwoPoint) # mate:交叉
toolbox.register("mutate", tools.mutGaussian, mu=0, sigma=1, indpb=0.1) # mutate : 变异
toolbox.register("select", tools.selTournament, tournsize=3) # select : 选择保留的最佳个体
toolbox.register("evaluate", evaluate)  # commit our evaluate


# Algorithms
def main():
    # create an initial population of 300 individuals (where
    # each individual is a list of integers)
    pop = toolbox.population(n=50)
    CXPB, MUTPB, NGEN = 0.5, 0.2, 40

    '''
    # CXPB  is the probability with which two individuals
    #       are crossed
    #
    # MUTPB is the probability for mutating an individual
    #
    # NGEN  is the number of generations for which the
    #       evolution runs
    '''

    # Evaluate the entire population
    fitnesses = map(toolbox.evaluate, pop)
    for ind, fit in zip(pop, fitnesses):
        ind.fitness.values = fit

    print("  Evaluated %i individuals" % len(pop))  # 这时候，pop的长度还是300呢
    print("-- Iterative %i times --" % NGEN)

    for g in range(NGEN):
        if g % 10 == 0:
            print("-- Generation %i --" % g)
        # Select the next generation individuals
        offspring = toolbox.select(pop, len(pop))
        # Clone the selected individuals
        offspring = list(map(toolbox.clone, offspring))
        # Change map to list,The documentation on the official website is wrong

        # Apply crossover and mutation on the offspring
        for child1, child2 in zip(offspring[::2], offspring[1::2]):
            if random.random() < CXPB:
                toolbox.mate(child1, child2)
                del child1.fitness.values
                del child2.fitness.values

        for mutant in offspring:
            if random.random() < MUTPB:
                toolbox.mutate(mutant)
                del mutant.fitness.values

        # Evaluate the individuals with an invalid fitness
        invalid_ind = [ind for ind in offspring if not ind.fitness.valid]
        fitnesses = map(toolbox.evaluate, invalid_ind)
        for ind, fit in zip(invalid_ind, fitnesses):
            ind.fitness.values = fit

        # The population is entirely replaced by the offspring
        pop[:] = offspring

    print("-- End of (successful) evolution --")

    best_ind = tools.selBest(pop, 1)[0]

    return best_ind, best_ind.fitness.values  # return the result:Last individual,The Return of Evaluate function


if __name__ == "__main__":
    # t1 = time.clock()
    best_ind, best_ind.fitness.values = main()
    # print(pop, best_ind, best_ind.fitness.values)
    # print("pop",pop)
    print("best_ind",best_ind)
    print("best_ind.fitness.values",best_ind.fitness.values)

    # t2 = time.clock()

    # print(t2-t1)