ColoracaoGrafos_IAnOTM_pt1.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Algoritmo de Coloração de Grafos | Inteligência Artificial¹ & Otimização I²\n",
    "##### Arthur Severo¹, Vitor Raposo¹, Victor Le Roy¹, Julio Cesar¹², Marco Túlio¹², Luisa Oliveira²"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Implementação do Algoritimo Genético"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Observações importantes:\n",
    "* As cores serão representadas por números\n",
    "* Podemos gerar grafos manualmente e automaticamente"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.utils import print_population, generateGraph\n",
    "from src.manual_graph import manual_graph_population_img1\n",
    "\n",
    "import random\n",
    "import copy\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "''' VARIAVEIS GLOBAIS PRE DEFINIDAS PARA EXECUCAO '''\n",
    "\n",
    "# Definicoes basicas do algoritmo\n",
    "MAX_GENERATIONS = 1000\n",
    "POP_SIZE = 10\n",
    "MUTATION_RATE = 0.1\n",
    "\n",
    "# Definicoes das cores\n",
    "MAX_NUM_COLORS = 3\n",
    "\n",
    "# Definicoes do grafo\n",
    "GRAPH_AUTO_GEN = False\n",
    "AUTO_GEN_GRAPH_SIZE = 10"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image](./image/example1.png)\n",
    "<p>Imagem 1</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_individual(graph):\n",
    "    '''\n",
    "        O individuo eh definido por um conjunto de vertices com cores aleatorias\n",
    "    '''\n",
    "    individual = graph\n",
    "    for valor_vertice, vertice in individual.vertices.items():\n",
    "        vertice.color = np.random.randint(1, MAX_NUM_COLORS) \n",
    "    return individual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fitness(individual):\n",
    "    '''\n",
    "        Fitness vai ser a quantidade de arestas que tem a mesma cor,\n",
    "        quanto menor o fitness, melhor a solucao\n",
    "    '''\n",
    "    fitness = 0\n",
    "    for i in range(1, len(individual.vertices)+1):\n",
    "        for j in range(i, len(individual.vertices)+1):\n",
    "            if ((individual.vertices[i].color == individual.vertices[j].color) and (individual.vertices[j] in individual.vertices[i].adjacencias.values())):\n",
    "                fitness += 1\n",
    "    return fitness"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mutation (individual):  \n",
    "    '''\n",
    "        A mutacao gera um numero aleatorio, verifica se este numero eh menor que o valor\n",
    "        da taxa de mutacao e, caso verdadeiro, altera a cor do vertice. Esse procedimento\n",
    "        ocorre para todos os vertices dentro de um individuo\n",
    "    '''\n",
    "    for i in range(1, len(individual.vertices)+1):\n",
    "        if np.random.rand() < MUTATION_RATE:\n",
    "            individual.vertices[i].color = np.random.randint(1, MAX_NUM_COLORS)\n",
    "    return individual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def crossover(graph, parent1, parent2):\n",
    "    '''\n",
    "        Crossover de ponto:\n",
    "        Eh gerado um ponto aleatorio que sera considerado a posicao onde a populacao sera dividida\n",
    "    '''\n",
    "    position = random.randint(2, len(graph.vertices)-2)\n",
    "    child1 = []\n",
    "    child2 = []\n",
    "    for i in range(position+1):\n",
    "        child1.append(parent1[i])\n",
    "        child2.append(parent2[i])\n",
    "    for i in range(position+1, len(graph.vertices)):\n",
    "        child1.append(parent2[i])\n",
    "        child2.append(parent1[i])\n",
    "    return child1, child2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sort_population(population):\n",
    "    return sorted(population, key=lambda individual: individual.fitness)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "individuo 1:\n",
      "Grafo:\n",
      "Fitness: 5     Mapped Fitness: None\n",
      "\n",
      "individuo 2:\n",
      "Grafo:\n",
      "Fitness: 6     Mapped Fitness: None\n",
      "\n",
      "individuo 3:\n",
      "Grafo:\n",
      "Fitness: 6     Mapped Fitness: None\n",
      "\n",
      "individuo 4:\n",
      "Grafo:\n",
      "Fitness: 7     Mapped Fitness: None\n",
      "\n",
      "individuo 5:\n",
      "Grafo:\n",
      "Fitness: 7     Mapped Fitness: None\n",
      "\n",
      "individuo 6:\n",
      "Grafo:\n",
      "Fitness: 8     Mapped Fitness: None\n",
      "\n",
      "individuo 7:\n",
      "Grafo:\n",
      "Fitness: 8     Mapped Fitness: None\n",
      "\n",
      "individuo 8:\n",
      "Grafo:\n",
      "Fitness: 8     Mapped Fitness: None\n",
      "\n",
      "individuo 9:\n",
      "Grafo:\n",
      "Fitness: 8     Mapped Fitness: None\n",
      "\n",
      "individuo 10:\n",
      "Grafo:\n",
      "Fitness: 11     Mapped Fitness: None\n",
      "\n"
     ]
    }
   ],
   "source": [
    "graph = generateGraph(GRAPH_AUTO_GEN, AUTO_GEN_GRAPH_SIZE, manual_graph_population_img1)\n",
    "# graph.print_graph()\n",
    "\n",
    "''' \n",
    "    * Criando a populacao inicial (populacao de grafos)\n",
    "    * Cada individuo da populacao eh um grafo contendo uma cor para cada vertice\n",
    "'''\n",
    "pop_size = POP_SIZE\n",
    "population = []\n",
    "for i in range(pop_size):\n",
    "    graph_copy = copy.deepcopy(graph)\n",
    "    individual = create_individual(graph_copy)\n",
    "    population.append(individual)\n",
    "for i in population:\n",
    "    i.setFitness(fitness, i)\n",
    "population = sort_population(population)\n",
    "print_population(population, False)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import Graph from Adj Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "csv:\n",
      "\t[['#658cda', '0', '1', '0', '0', '0'], ['#cac707', '0', '0', '1', '1', '0'], ['#62e1a3', '0', '1', '0', '0', '1'], ['#8b2cb0', '1', '0', '0', '0', '0'], ['#eca34a', '0', '1', '0', '0', '0']]\n",
      "Nome: a     Valor: 0     Cor: #658cda     \n",
      "Nome: b     Valor: 1     Cor: #cac707     \n",
      "Nome: c     Valor: 2     Cor: #62e1a3     \n",
      "Nome: d     Valor: 3     Cor: #8b2cb0     \n",
      "Nome: e     Valor: 4     Cor: #eca34a     \n",
      "add 0->1: 1\n",
      "add 1->2: 1\n",
      "add 1->3: 1\n",
      "add 2->1: 1\n",
      "add 2->4: 1\n",
      "add 3->0: 1\n",
      "add 4->1: 1\n",
      "Grafo:\n",
      "Fitness: None     Mapped Fitness: None\n",
      "    vertice: \n",
      "        Nome: a     Valor: 0     Cor: #658cda     \n",
      "     vizinhos: \n",
      "        Nome: b     Valor: 1     Cor: #cac707     \n",
      "        Nome: d     Valor: 3     Cor: #8b2cb0     \n",
      "\n",
      "    vertice: \n",
      "        Nome: b     Valor: 1     Cor: #cac707     \n",
      "     vizinhos: \n",
      "        Nome: a     Valor: 0     Cor: #658cda     \n",
      "        Nome: c     Valor: 2     Cor: #62e1a3     \n",
      "        Nome: d     Valor: 3     Cor: #8b2cb0     \n",
      "        Nome: e     Valor: 4     Cor: #eca34a     \n",
      "\n",
      "    vertice: \n",
      "        Nome: c     Valor: 2     Cor: #62e1a3     \n",
      "     vizinhos: \n",
      "        Nome: b     Valor: 1     Cor: #cac707     \n",
      "        Nome: e     Valor: 4     Cor: #eca34a     \n",
      "\n",
      "    vertice: \n",
      "        Nome: d     Valor: 3     Cor: #8b2cb0     \n",
      "     vizinhos: \n",
      "        Nome: b     Valor: 1     Cor: #cac707     \n",
      "        Nome: a     Valor: 0     Cor: #658cda     \n",
      "\n",
      "    vertice: \n",
      "        Nome: e     Valor: 4     Cor: #eca34a     \n",
      "     vizinhos: \n",
      "        Nome: c     Valor: 2     Cor: #62e1a3     \n",
      "        Nome: b     Valor: 1     Cor: #cac707     \n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from src.auto_graph import auto_graph_readfile\n",
    "\n",
    "path = \"./data/adjmtx1.csv\"\n",
    "g = auto_graph_readfile(path)\n",
    "g.print_graph()\n",
    "g.visualize()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.5 64-bit",
   "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": "3.10.5 (tags/v3.10.5:f377153, Jun  6 2022, 16:14:13) [MSC v.1929 64 bit (AMD64)]"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "bebf9730dc43cfbc35fb329a82a69476f49cde6de038a2bbfc26f12404ad85fb"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}