Revisi_GA_MF_Tournament.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.7.4-final"
    },
    "colab": {
      "name": "Revisi GA_MF Tournament.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5f6mVI5AenuM"
      },
      "source": [
        "### Genetic Algorithm"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "neoxLdKZenuO"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import math\n",
        "import copy\n",
        "import random"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oy2x3DdbenuX"
      },
      "source": [
        "uk_populasi = 20\n",
        "uktour = 5\n",
        "ukkrom = 10\n",
        "pc = 0.76\n",
        "pm = 0.31"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3RoojYdIenug"
      },
      "source": [
        "**Inisiasi Populasi Awal**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u8K1ZwWOenuh"
      },
      "source": [
        "def generatePopulasi(uk_pop, uk_krom):\n",
        "    pop = []\n",
        "    for i in range(uk_pop):\n",
        "        pop.append([random.randint(0,9) for i in range(uk_krom)])\n",
        "    return pop"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fNcprspVTnIP"
      },
      "source": [
        "**Decode**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZH_KSyiJenun"
      },
      "source": [
        "def phenotype(krom):\n",
        "    x_x1 = 0\n",
        "    d_x1 = 0\n",
        "    x_x2 = 0\n",
        "    d_x2 = 0\n",
        "\n",
        "    j = 1\n",
        "    k = 1\n",
        "    for i in range (len(krom)):\n",
        "        if (i < (len(krom) / 2)):\n",
        "            x_x1 += (krom[i] * (1 / np.power(10,j)))\n",
        "            d_x1 += (9 / np.power(10,j))\n",
        "            j+=1\n",
        "        else:\n",
        "            x_x2 += (krom[i] * (1 / np.power(10,k)))\n",
        "            d_x2 += (9/ np.power(10,k))\n",
        "            k+=1\n",
        "    x1 = -1 + (3 * x_x1 / d_x1) #phenotype x1\n",
        "    x2 = -1 + (2 * x_x2 / d_x2) #phenotype x2\n",
        "\n",
        "    return [x1,x2]"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I7Wk9SJAXCom"
      },
      "source": [
        "**Hitung Fitness**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R8TS9V9jenuw"
      },
      "source": [
        "def hitungFitness(krom):\n",
        "    k = phenotype(krom)\n",
        "    f = -(math.cos(k[0]) * math.sin(k[1]) - (k[0]/((k[1])**2 + 1)))\n",
        "    return f"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zYvEyuI4LlPc"
      },
      "source": [
        "def hitungFitnessAll(pop):\n",
        "    fit_all = []    \n",
        "    for i in range(len(pop)):\n",
        "        fit_all.append(hitungFitness(pop[i]))\n",
        "    return fit_all"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lfcdk3FLenuy"
      },
      "source": [
        "**Parent Selection**\n",
        "\n",
        "disini saya akan menggunakan Tournament"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IZAmS5Pcenu0"
      },
      "source": [
        "def tournamentSelection(pop, uk_tour):\n",
        "    best_krom = []\n",
        "    for i in range(1, uk_tour):\n",
        "        krom = pop[random.randint(0,len(pop)-1)]\n",
        "        if (best_krom == [] or hitungFitness(krom) > hitungFitness(best_krom)):\n",
        "            best_krom = krom\n",
        "    return best_krom"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a-tTrMeqenu2"
      },
      "source": [
        "**Crossover**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "01YTsJneenu2"
      },
      "source": [
        "def crossover(par1, par2, pc, ukrom):\n",
        "    r = random.random()\n",
        "    if (r < pc):\n",
        "        point = random.randint(0,ukrom-1)\n",
        "        for i in range(point):\n",
        "            temp1 = par2[i]\n",
        "            temp2 = par1[i]\n",
        "            par1[i] = temp1\n",
        "            par2[i] = temp2\n",
        "    return par1, par2"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lY6FfgHAenu3"
      },
      "source": [
        "**Mutasi**\n",
        "\n",
        "melakukan mutasi apabila nilai random lebih kecil dari probabilitas mutasi"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jgvVQJEqE2bj"
      },
      "source": [
        "def mutasi(par1, par2, pm, uk_krom):\n",
        "    r = random.random()\n",
        "    if (r < pm):\n",
        "        par1[random.randint(0,uk_krom-1)] = random.randint(0,9)\n",
        "        par2[random.randint(0,uk_krom-1)] = random.randint(0,9)\n",
        "    return par1, par2"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4TqDqJGXenu6"
      },
      "source": [
        "**Pergantian Generasi**\n",
        "\n",
        "berdasarkan fitness"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SC0yRTWTenu6"
      },
      "source": [
        "def getElitisme(fit_all):\n",
        "    return fit_all.index(max(fit_all))"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3f5zetA7enu7"
      },
      "source": [
        "***Main Program***"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "tags": [],
        "id": "HxhLAUFUenu8",
        "outputId": "77ceff6f-847d-449e-c1c6-aa8dda0757c0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 415
        }
      },
      "source": [
        "populasi = generatePopulasi(uk_populasi, ukkrom)\n",
        "fit = []\n",
        "generasi = 100\n",
        "\n",
        "for i in range(generasi):\n",
        "    fitness = hitungFitnessAll(populasi)\n",
        "    new_populasi = []\n",
        "\n",
        "    best = getElitisme(fitness)\n",
        "    fit.append(hitungFitness(populasi[best]))\n",
        "    new_populasi.append(populasi[best])\n",
        "    new_populasi.append(populasi[best])\n",
        "    i = 0\n",
        "    while (i < uk_populasi-2):\n",
        "        parent1 = tournamentSelection(populasi, uktour)\n",
        "        parent2 = tournamentSelection(populasi, uktour)\n",
        "        while (parent1 == parent2):\n",
        "            parent2 = tournamentSelection(populasi, uktour)\n",
        "        par1, par2 = copy.deepcopy(parent1), copy.deepcopy(parent2)\n",
        "        child = crossover(par1, par2, pc, ukkrom)\n",
        "        child = mutasi(child[0], child[1], pm, ukkrom)\n",
        "        new_populasi += child\n",
        "        i += 2\n",
        "    populasi = new_populasi\n",
        "\n",
        "fitness = hitungFitnessAll(populasi)\n",
        "result = getElitisme(fitness)\n",
        "\n",
        "print('Result')\n",
        "print()\n",
        "print('Best Chromosom/Induvidu  = ', populasi[result])\n",
        "print('Best Fitness             = ', hitungFitness(populasi[result]))\n",
        "print('Decode Result            = ', phenotype(populasi[result]))\n",
        "plt.plot(fit)\n",
        "plt.suptitle(\"Generasi Fitness\", size=18)\n",
        "plt.xlabel('Generasi', fontsize=16)\n",
        "plt.ylabel('Fitness', fontsize=16)\n",
        "plt.show"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Result\n",
            "\n",
            "Best Chromosom/Induvidu  =  [9, 9, 9, 9, 9, 5, 5, 2, 9, 0]\n",
            "Best Fitness             =  2.0218067819202177\n",
            "Decode Result            =  [2.0, 0.10581105811058111]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<function matplotlib.pyplot.show>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iIhQ0BARUsAp"
      },
      "source": [
        "**Thank You**"
      ]
    }
  ]
}