{ "cells": [ { "cell_type": "markdown", "id": "dd2ed0d5-ee09-452b-ace5-89b2be772bb8", "metadata": {}, "source": [ "# Usando algorítmo genético\n", "### Encontrar o máximo de $f(x)=cos(20x) - \\frac{|x|}{2} + \\frac{x^3}{4}$, $-2 \\le x \\le 2$" ] }, { "cell_type": "markdown", "id": "0f0f670a-282f-4f71-824a-33ca5be6e9b4", "metadata": {}, "source": [ "# Importando pacotes necessários" ] }, { "cell_type": "code", "execution_count": 1, "id": "8d5900cb", "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "markdown", "id": "950a2db7-4386-4538-9f13-04d3107d7675", "metadata": {}, "source": [ "### Gráfico" ] }, { "cell_type": "code", "execution_count": 2, "id": "73c001a9-ee73-4399-9223-978994f6c2ce", "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "x=np.arange(-2,2.01,0.01) # gera x de -2 até 2 com passo 0.01\n", "y=np.cos(20*x)-np.abs(x)/2+np.power(x,3)/4\n", "plt.plot(x,y)\n", "plt.grid('on')" ] }, { "cell_type": "markdown", "id": "af30d0cb-4fe2-4d02-a464-8893f1bad7fb", "metadata": {}, "source": [ "# Definindo variáveis" ] }, { "cell_type": "markdown", "id": "5136df45-d732-4bbf-bcdc-ccf703f0afad", "metadata": {}, "source": [ "### Tamanho do cromossomo" ] }, { "cell_type": "code", "execution_count": 3, "id": "a591da6d-011b-4779-95c1-5327880eb768", "metadata": {}, "outputs": [], "source": [ "nBits=22 #número de genes" ] }, { "cell_type": "markdown", "id": "4367c004-016c-4e73-9b3c-f26086f7d6d4", "metadata": {}, "source": [ "### Tamanho da população" ] }, { "cell_type": "code", "execution_count": 4, "id": "f2035fb3-73cc-4afd-a85e-43658d1ca4ee", "metadata": {}, "outputs": [], "source": [ "nPop=20" ] }, { "cell_type": "markdown", "id": "e1760ab2-5870-41ec-a3ef-db1ae2e87fdd", "metadata": {}, "source": [ "### Probabilidade de mutação" ] }, { "cell_type": "code", "execution_count": 5, "id": "1e71a657-88e5-42be-ac68-9701f6d18683", "metadata": {}, "outputs": [], "source": [ "pMut=0.15 #15% de probabilidade de mutar" ] }, { "cell_type": "markdown", "id": "0898324a-1f7a-4606-b588-3c9477616ed1", "metadata": {}, "source": [ "### Convergência" ] }, { "cell_type": "markdown", "id": "efb646ce-bb26-41f1-81a4-89186e957764", "metadata": {}, "source": [ "#### Intervalo de convergência" ] }, { "cell_type": "code", "execution_count": 6, "id": "761b3b8c-1a90-49db-90f2-f986a73637bd", "metadata": {}, "outputs": [], "source": [ "intConvergencia=0.0001" ] }, { "cell_type": "markdown", "id": "c2935207-359d-483b-8d2c-e17e8ed29bc2", "metadata": {}, "source": [ "#### Número de gerações com solução próxima da anterior" ] }, { "cell_type": "code", "execution_count": 7, "id": "41cbc6f5-1692-4e96-b07e-843f442e133d", "metadata": {}, "outputs": [], "source": [ "nGeracoesConv = 10" ] }, { "cell_type": "markdown", "id": "8b660eb1-61eb-42ec-a442-665b76220f14", "metadata": {}, "source": [ "### Intervalo da solução" ] }, { "cell_type": "code", "execution_count": 8, "id": "f1079e32-7cd1-46e0-a93a-29e7df910fa9", "metadata": {}, "outputs": [], "source": [ "vMin=-2\n", "vMax=2" ] }, { "cell_type": "markdown", "id": "866db291-5b21-47bc-9827-68061fe45a13", "metadata": {}, "source": [ "### Gerando uma população inicial" ] }, { "cell_type": "code", "execution_count": 9, "id": "6f860c5d", "metadata": {}, "outputs": [], "source": [ "cromossomo=[] #vários cromossomos, vários indivíduos ou soluções\n", "acromossomo = np.random.randint(0,2,(nPop,nBits))#array de 22 bits cada\n", "for i in range(nPop):\n", " cromossomo.append(''.join(str(x) for x in acromossomo[i]))\n", "\n", "#processo semelhante ao acima com array \n", "pot=np.power(2,np.arange(nBits-1,-1,-1))#potências de 2²¹ a 2⁰\n", "#acromossomo" ] }, { "cell_type": "code", "execution_count": 10, "id": "c3419b1d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[1, 1, 1, 1, 0, 1, 0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0],\n", " [0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 1, 1, 0, 0, 0, 1, 1, 0, 0, 1, 1],\n", " [0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0],\n", " [1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 0, 0, 0, 1, 0, 0, 1, 1, 0, 1],\n", " [0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 0, 1, 1, 0, 1, 1, 1, 0],\n", " [1, 0, 1, 1, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0],\n", " [0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1],\n", " [1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0],\n", " [1, 0, 0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 1, 1, 1, 1, 1, 0],\n", " [1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0],\n", " [0, 0, 1, 0, 0, 1, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 0, 0, 1],\n", " [0, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 0, 0, 1, 1, 0, 0, 1, 1, 0],\n", " [1, 0, 0, 1, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0],\n", " [1, 1, 1, 1, 1, 0, 1, 0, 1, 0, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 1, 1],\n", " [0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1],\n", " [0, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1],\n", " [1, 1, 0, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0],\n", " [1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 1, 0, 1, 0, 1, 0, 0, 1, 0],\n", " [1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 1],\n", " [0, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 1]])" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "acromossomo" ] }, { "cell_type": "markdown", "id": "28faca9d-510e-4750-9559-5b872e3cc10d", "metadata": {}, "source": [ "### População inicial gerada" ] }, { "cell_type": "code", "execution_count": 11, "id": "0cb26868-1973-4c91-bbf6-334ca4704d30", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['1111010101111100000110',\n", " '0101110101011000110011',\n", " '0001101000101100000010',\n", " '1110011011100001001101',\n", " '0110001111000101101110',\n", " '1011011010010000011010',\n", " '0010110000000001111111',\n", " '1111010110110101111000',\n", " '1001111010010010111110',\n", " '1000010111111011100110',\n", " '0010010011111011010001',\n", " '0111110110101001100110',\n", " '1001101010001010101110',\n", " '1111101010110101100011',\n", " '0011101111011010011111',\n", " '0011101110111011111001',\n", " '1101011101111111000000',\n", " '1010010010010101010010',\n", " '1111111010011010000101',\n", " '0011101110100001100011']" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cromossomo" ] }, { "cell_type": "markdown", "id": "dc5732ff-4608-4e03-bc9f-863e3a816bb6", "metadata": {}, "source": [ "### Converter sequência de bits em reais decimais, conforme os valores máximos e mínimos do intervalo de soluções $-2\\le x \\le2$\n", "### $$V_r = V_{min} + (V_{max} - V_{min})\\frac{V_{dec}}{2^g - 1}$$\n", "### onde\n", "### $V_r$: valor em real, que será dentro do intervalo $[V_{min}, V_{max}]$, neste caso $[-2,2]$\n", "### $V_{min}$: valor mínimo do intervalo\n", "### $V_{max}$: valor máximo \n", "### $V_{dec}$: valor decimal do cromossomo\n", "### $g$: número de genes (bits) do cromossomo" ] }, { "cell_type": "code", "execution_count": 13, "id": "f9de50c1-6e1f-47a7-95d8-0c411239a858", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([ 1.8357 , -0.54145492, -1.59106245, 1.60749617, -0.44105683,\n", " 0.85256454, -1.31237872, 1.83922668, 0.47772085, 0.09348156,\n", " -1.42216383, -0.03652335, 0.4147173 , 1.91733168, -1.06478955,\n", " -1.06665684, 1.36712727, 0.57161202, 1.97815513, -1.06826474])" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#transformando vetor de bits e reais no intervalo [vMin, vMax]\n", "dec=acromossomo@pot\n", "axr=vMin + (vMax-vMin)*dec/(2**nBits - 1)\n", "axr" ] }, { "cell_type": "markdown", "id": "6318f16b", "metadata": {}, "source": [ "### O código da célula abaixo faz o mesmo que o da célula [12]\n", "#### Pode escolher um dos códigos" ] }, { "cell_type": "code", "execution_count": 14, "id": "16bba39e", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[ 1.8357 -0.54145492 -1.59106245 1.60749617 -0.44105683 0.85256454\n", " -1.31237872 1.83922668 0.47772085 0.09348156 -1.42216383 -0.03652335\n", " 0.4147173 1.91733168 -1.06478955 -1.06665684 1.36712727 0.57161202\n", " 1.97815513 -1.06826474]\n" ] } ], "source": [ "#função que poderia ser utilizada para fazer o mesmo da célula [12]\n", "def bin2float(b):\n", " dec=int(b,2) #b é uma string de 0s e 1s int('1100',2) retorna 12\n", " return vMin + (vMax-vMin)*dec/(2**nBits - 1)\n", "\n", "xr=np.array(list(map(bin2float,cromossomo))) \n", "print(xr)" ] }, { "cell_type": "markdown", "id": "c1f8aab8-7658-4455-ac3c-645fd68170ef", "metadata": {}, "source": [ "### $$f(x)$$" ] }, { "cell_type": "markdown", "id": "d89e8cfe", "metadata": {}, "source": [ "## Gerando os 20 valores de $f(x)$" ] }, { "cell_type": "code", "execution_count": 16, "id": "a7a82839-6135-4f6c-9490-c5611904d4ad", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([ 1.18140278, -0.47612048, -0.8834915 , 0.97721519, -1.06525236,\n", " -0.49686945, -0.78095198, 1.24592474, -1.20321301, -0.34094353,\n", " -2.41594436, 0.72658903, -0.61579583, 1.60102023, -1.60202785,\n", " -1.62793718, -0.64116009, 0.18380688, 0.6570892 , -1.64936878])" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fx=np.cos(20*xr)-np.abs(xr)/2+np.power(xr,3)/4\n", "fx" ] }, { "cell_type": "markdown", "id": "630c359e-84da-4caf-a8b2-9d3ea61a3c50", "metadata": {}, "source": [ "### A função de fitness é $f(x)+4$, assim retiramos os valores negativos para poderemos ter a função de distribuição acumulada" ] }, { "cell_type": "code", "execution_count": 18, "id": "349975f1", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([5.18140278, 3.52387952, 3.1165085 , 4.97721519, 2.93474764,\n", " 3.50313055, 3.21904802, 5.24592474, 2.79678699, 3.65905647,\n", " 1.58405564, 4.72658903, 3.38420417, 5.60102023, 2.39797215,\n", " 2.37206282, 3.35883991, 4.18380688, 4.6570892 , 2.35063122])" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fitness = fx+4\n", "fitness" ] }, { "cell_type": "markdown", "id": "07c60c43-210a-4a3d-ae67-a22db0fd6935", "metadata": {}, "source": [ "### Calculando a probabilidade de cada elemento $x$ ser a solução" ] }, { "cell_type": "code", "execution_count": 19, "id": "8df5924e-8441-4704-9d49-c1f4c5570f04", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([0.07119857, 0.04842225, 0.04282449, 0.06839279, 0.04032689,\n", " 0.04813714, 0.04423351, 0.07208518, 0.03843114, 0.05027974,\n", " 0.02176679, 0.06494889, 0.04650295, 0.07696461, 0.03295096,\n", " 0.03259493, 0.04615441, 0.05749043, 0.06399389, 0.03230044])" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "solucaoAnterior=fx[np.argmin(fitness)] \n", "novaSolucao=fx[np.argmax(fitness)]\n", "numConv=0\n", "if abs(novaSolucao - solucaoAnterior) <= intConvergencia:\n", " numConv += 1\n", "#while numConvr)\n", " sel = cromossomo[id] \n", " if i>0:\n", " while sel in pais: #evita seleção repetida\n", " r=np.random.uniform()\n", " id=np.argmax(prbAc>r)\n", " sel=cromossomo[id]\n", " pais.append(sel)\n", " \n", "pais" ] }, { "cell_type": "markdown", "id": "13c2a595", "metadata": {}, "source": [ "### Lembrando que poderia selecionar os dez maiores valores\n", "#### Mas foi escolhido por roleta (acaso)" ] }, { "cell_type": "markdown", "id": "c4357e4c-9dfb-4ddc-bfd7-2e4ac9a366a3", "metadata": {}, "source": [ "### Cruzamento" ] }, { "cell_type": "code", "execution_count": 22, "id": "ceaeaf90", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['0010110000001010101110',\n", " '1001101010000001111111',\n", " '0111110110001010101110',\n", " '1001101010101001100110',\n", " '1111111010001010101110',\n", " '1001101010011010000101',\n", " '1111101010111010101110',\n", " '1001101010000101100011',\n", " '0011101110001010101110',\n", " '1001101011011010011111',\n", " '1111010101001010101110',\n", " '1001101010111100000110',\n", " '1010101010001010101110',\n", " '1001010010010101010010',\n", " '1111010110110110101110',\n", " '1001101010001001111000',\n", " '0011101110100010101110',\n", " '1001101010001001100011',\n", " '0111110110101001111111',\n", " '0010110000000001100110',\n", " '1111111010011001111111',\n", " '0010110000000010000101',\n", " '1111101010110101100111',\n", " '0010110000000001111011',\n", " '0011110000000001111111',\n", " '0010101111011010011111',\n", " '1111010000000001111111',\n", " '0010110101111100000110',\n", " '1010110000000001111111',\n", " '0010010010010101010010',\n", " '1111010110000001111111',\n", " '0010110000110101111000',\n", " '0011101110100001111111',\n", " '0010110000000001100011',\n", " '1111111010001001100110',\n", " '0111110110111010000101',\n", " '1111110110101001100110',\n", " '0111101010110101100011',\n", " '0011110110101001100110',\n", " '0111101111011010011111',\n", " '1111010101111100100110',\n", " '0111110110101001000110',\n", " '1010010010010101010110',\n", " '0111110110101001100010',\n", " '1111010110101001100110',\n", " '0111110110110101111000',\n", " '0011101110100001100110',\n", " '0111110110101001100011',\n", " '1111101010011010000101',\n", " '1111111010110101100011',\n", " '0011111010011010000101',\n", " '1111101111011010011111',\n", " '1111011010011010000101',\n", " '1111110101111100000110',\n", " '1010010010010010000101',\n", " '1111111010011101010010',\n", " '1111010110111010000101',\n", " '1111111010010101111000',\n", " '0011101110100000000101',\n", " '1111111010011011100011',\n", " '0011101111011001100011',\n", " '1111101010110110011111',\n", " '1111010101111100000011',\n", " '1111101010110101100110',\n", " '1010010010010101100011',\n", " '1111101010110101010010',\n", " '1111010110110101100011',\n", " '1111101010110101111000',\n", " '0011101010110101100011',\n", " '1111101110100001100011',\n", " '1111010111011010011111',\n", " '0011101101111100000110',\n", " '1010010010011010011111',\n", " '0011101111010101010010',\n", " '1111010110110101111111',\n", " '0011101111011010011000',\n", " '0011101110100001011111',\n", " '0011101111011010100011',\n", " '1010010101111100000110',\n", " '1111010010010101010010',\n", " '1111010110110101111010',\n", " '1111010101111100000100',\n", " '0011110101111100000110',\n", " '1111001110100001100011',\n", " '1111010110110101010010',\n", " '1010010010010101111000',\n", " '0011101110100001110010',\n", " '1010010010010101000011',\n", " '0011101110100001101000',\n", " '1111010110110101110011']" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "filhos=[]\n", "for p1 in range(len(pais)-1):\n", " for p2 in range(p1+1,len(pais)):\n", " c=np.random.randint(1,20) #ponto de corte\n", " #Trabalhando com array numpy \n", " #afilho1[0:c+1]=apais[p2][0:c+1]\n", " #afilho1[c+1:]=apais[p1][c+1:]\n", " #afilho2[0:c+1]=apais[p1][0:c+1]\n", " #afilho2[c+1:]=apais[p2][c+1:]\n", " \n", " #Usando lista gasta menos linhas\n", " filho1=pais[p2][0:c+1]+pais[p1][c+1:]\n", " filho2=pais[p1][0:c+1]+pais[p2][c+1:]\n", " \n", " while (filho1 in filhos) or (filho2 in filhos) or (filho1 in cromossomo) or (filho2 in cromossomo):\n", " c=np.random.randint(1,20)\n", " filho1=pais[p2][0:c+1]+pais[p1][c+1:] #concatenando, são caracteres\n", " filho2=pais[p1][0:c+1]+pais[p2][c+1:]\n", " \n", " filhos.append(filho1)\n", " filhos.append(filho2) \n", "filhos" ] }, { "cell_type": "markdown", "id": "1130b4e6-4c2c-4b87-8347-43a704822573", "metadata": {}, "source": [ "### Mutação\n", "##### O gene a ser mutado aparece entre [ ]" ] }, { "cell_type": "code", "execution_count": 23, "id": "a1f09714", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Mutou filho 011111 [0] 110001010101110\n", "Para filho 011111 [1] 110001010101110\n", "Mutou filho 10011010101111000001 [1] 0\n", "Para filho 10011010101111000001 [0] 0\n", "Mutou filho [1] 001101010001001111000\n", "Para filho [0] 001101010001001111000\n", "Mutou filho 01111101 [1] 0101001111111\n", "Para filho 01111101 [0] 0101001111111\n", "Mutou filho 001111 [0] 000000001111111\n", "Para filho 001111 [1] 000000001111111\n", "Mutou filho 00100100100101010100 [1] 0\n", "Para filho 00100100100101010100 [0] 0\n", "Mutou filho 00101100000000 [0] 1100011\n", "Para filho 00101100000000 [1] 1100011\n", "Mutou filho 11 [1] 1010101111100100110\n", "Para filho 11 [0] 1010101111100100110\n", "Mutou filho 0111110110 [1] 01001000110\n", "Para filho 0111110110 [0] 01001000110\n", "Mutou filho 11111 [1] 1010010101111000\n", "Para filho 11111 [0] 1010010101111000\n", "Mutou filho 111101 [0] 110110101100011\n", "Para filho 111101 [1] 110110101100011\n", "Mutou filho 0011101101111100000 [1] 10\n", "Para filho 0011101101111100000 [0] 10\n", "Mutou filho 1 [1] 11010110110101111111\n", "Para filho 1 [0] 11010110110101111111\n" ] } ], "source": [ "#for i,f in enumerate(filhos):\n", "# r=random.uniform()\n", "# if r" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "x=np.arange(-2,2.01,0.01)\n", "y=np.cos(20*x)-np.abs(x)/2+np.power(x,3)/4\n", "plt.plot(x,y)\n", "plt.plot(maxX,maxF,'ro')\n", "plt.grid('on')" ] }, { "cell_type": "markdown", "id": "8ae6d5c1-abd2-4c90-8d00-484c7d841978", "metadata": {}, "source": [ "# A seleção natural pode ser feita de várias maneiras, tais como:\n", "### 1) Tomando os 20 melhores individuos;\n", "### 2) Tomando aleatoriamente 19 indivíduos mais o melhor indivíduo.\n", "# O critério de parada pode ser, entre outros:\n", "### 1) Arbitrar um determinado número de iterações;\n", "### 2) Melhor solução nova não diferencie da anterior por um certo número de repetições;\n", "### 3) Melhor solução nova apresente pouca variação com a solução anterior." ] }, { "cell_type": "markdown", "id": "a3d1506d-1c2d-4097-9099-0178d21cb2f0", "metadata": {}, "source": [ "# Selecionando os indivíduos de maior fitness da nova geração" ] }, { "cell_type": "code", "execution_count": 54, "id": "72173719-577b-41d3-aa7d-cd4db107accf", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([ 62, 72, 47, 96, 89, 63, 38, 55, 5, 9, 100, 71, 76,\n", " 92, 32, 105, 16, 84, 80, 97])" ] }, "execution_count": 54, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ind = np.argsort(fitness)[-nPop:] #toma os nPop melhores indivíduos\n", "ind" ] }, { "cell_type": "code", "execution_count": 55, "id": "028d843f-6b83-417b-ba70-38a0478c600d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['0011101010001000101101',\n", " '0110111101010111110000',\n", " '1010010000010110100011',\n", " '1101010101101010100110',\n", " '1111000001110001001100',\n", " '0110101110111110010011',\n", " '0001001110010100100011',\n", " '0100010111000001101010',\n", " '1000010001100100001011',\n", " '1001010000001111010000',\n", " '1010010011111011001111',\n", " '1010000000101110110110',\n", " '0101010010101010100000',\n", " '0110011010101111110110',\n", " '0010100001001100111111',\n", " '1000100111001100101011',\n", " '1111101011001000101000',\n", " '0100101001110111100000',\n", " '0101111001110111000100',\n", " '1010010001001101111110']" ] }, "execution_count": 55, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ind = np.argsort(fx)[-nPop:]\n", "cromossomo=[]\n", "cromossomo = list(np.array(novageracao)[ind])\n", "cromossomo" ] }, { "cell_type": "code", "execution_count": 56, "id": "63ec31a9-884f-46ef-be78-c9d3b00bd902", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([-1.82259854, 0.33871644, 0.04112349, -0.06208825, -0.35877524,\n", " 0.83271344, -2.8423068 , 0.1535618 , 0.16297873, 0.85087617,\n", " 0.29145166, -1.0262422 , 0.14460895, -0.2695544 , -1.97452859,\n", " -1.07252363, 1.58938741, -1.08691601, -0.79129364, 0.10453972])" ] }, "execution_count": 56, "metadata": {}, "output_type": "execute_result" } ], "source": [ "xr=np.array(list(map(bin2float,cromossomo))) \n", "fx = np.cos(20*xr)-np.abs(xr)/2+np.power(xr,3)/4\n", "fx" ] }, { "cell_type": "code", "execution_count": 63, "id": "949c63a3-bd64-4417-9b06-78c669095e76", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1" ] }, "execution_count": 63, "metadata": {}, "output_type": "execute_result" } ], "source": [ "numConv=0\n", "novaSolucao=fx[-1]\n", "numConv += 1\n", "if abs(novaSolucao - solucao) > intConvergencia:\n", " numConv=0\n", "else:\n", " #break\n", "solucao=novaSolucao\n", "numConv" ] }, { "cell_type": "code", "execution_count": 90, "id": "b6ad6bfd", "metadata": {}, "outputs": [], "source": [ "def SubIndToChar(st,i,c):\n", " lst=list(st)\n", " lst[i]=c\n", " st=''.join(s for s in lst) \n", " \n", "#st[i]=c #erro" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.9" } }, "nbformat": 4, "nbformat_minor": 5 }