Agrupando_clientes_por_consumo_de_Energia_Mini_Projeto_.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "anaconda-cloud": {},
    "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"
    },
    "colab": {
      "name": "Agrupando clientes por consumo de Energia-Mini-Projeto-.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "nxBxqsO88Ywl"
      ],
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/luiz-star/Agrupamento_Clientes/blob/main/Agrupando_clientes_por_consumo_de_Energia_Mini_Projeto_.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9ApBP3PV8Yv8"
      },
      "source": [
        "# <font color='blue'> - Machine Learning</font>\n",
        "\n",
        "# <font color='blue'> - Agrupando Clientes Por Consumo de Energia</font>\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7KdxMQ3k8YwD"
      },
      "source": [
        "A partir de dados de consumo de energia de clientes, nosso trabalho é agrupar os consumidores por similaridade a afim de compreender o comportamento dos clientes e sua relação com o consumo de energia."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G9lD5iQI8YwF"
      },
      "source": [
        "Você deve executar as seguintes tarefas:\n",
        "    \n",
        "1- Tratar os valores ausentes nos dados.\n",
        "\n",
        "2- Coletar uma amostra de 1% dos dados para criar o modelo de clusterização com o K-Means.\n",
        "\n",
        "3- Encontrar o melhor valor de K para esse conjunto de dados.\n",
        "\n",
        "4- Criar o modelo K-Means usando o valor de K encontrado no item 3.\n",
        "\n",
        "5- Criar um Meshgrid para visualização de todos os clusters.\n",
        "\n",
        "6- Visualizar os centróides.\n",
        "\n",
        "7- Calcular o Silhoutte Score.\n",
        "\n",
        "8 - Calcular a média de consumo de energia por cluster (usar a coluna Global_active_power para o cálculo da média).\n",
        "\n",
        "Dica: Para o item 8, considere a amostra criada no item 2."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6kAxNME_8YwH"
      },
      "source": [
        "# Imports\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib import pylab\n",
        "from sklearn.cluster import KMeans\n",
        "from sklearn.decomposition import PCA\n",
        "from sklearn.model_selection import train_test_split\n",
        "from scipy.spatial.distance import cdist, pdist\n",
        "from sklearn.metrics import silhouette_score\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "%matplotlib inline"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qk8Mw0Tc8YwI"
      },
      "source": [
        "https://archive.ics.uci.edu/ml/datasets/Individual+household+electric+power+consumption"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Njg2G8wQ8YwJ"
      },
      "source": [
        "# Carregando os dados\n",
        "dataset = pd.read_csv('household_power_consumption.txt', delimiter = ';', low_memory = False)"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WBDOS8mT8YwK",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "f1a0d0a8-ca9e-4ba9-9cf0-f75da01a5108"
      },
      "source": [
        "dataset.head()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Date</th>\n",
              "      <th>Time</th>\n",
              "      <th>Global_active_power</th>\n",
              "      <th>Global_reactive_power</th>\n",
              "      <th>Voltage</th>\n",
              "      <th>Global_intensity</th>\n",
              "      <th>Sub_metering_1</th>\n",
              "      <th>Sub_metering_2</th>\n",
              "      <th>Sub_metering_3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>16/12/2006</td>\n",
              "      <td>17:24:00</td>\n",
              "      <td>4.216</td>\n",
              "      <td>0.418</td>\n",
              "      <td>234.840</td>\n",
              "      <td>18.400</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>16/12/2006</td>\n",
              "      <td>17:25:00</td>\n",
              "      <td>5.360</td>\n",
              "      <td>0.436</td>\n",
              "      <td>233.630</td>\n",
              "      <td>23.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>16.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>16/12/2006</td>\n",
              "      <td>17:26:00</td>\n",
              "      <td>5.374</td>\n",
              "      <td>0.498</td>\n",
              "      <td>233.290</td>\n",
              "      <td>23.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>16/12/2006</td>\n",
              "      <td>17:27:00</td>\n",
              "      <td>5.388</td>\n",
              "      <td>0.502</td>\n",
              "      <td>233.740</td>\n",
              "      <td>23.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>16/12/2006</td>\n",
              "      <td>17:28:00</td>\n",
              "      <td>3.666</td>\n",
              "      <td>0.528</td>\n",
              "      <td>235.680</td>\n",
              "      <td>15.800</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         Date      Time  ... Sub_metering_2 Sub_metering_3\n",
              "0  16/12/2006  17:24:00  ...          1.000           17.0\n",
              "1  16/12/2006  17:25:00  ...          1.000           16.0\n",
              "2  16/12/2006  17:26:00  ...          2.000           17.0\n",
              "3  16/12/2006  17:27:00  ...          1.000           17.0\n",
              "4  16/12/2006  17:28:00  ...          1.000           17.0\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dDG7qlQQ8YwN",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9e5c8ebd-4b25-42da-b8ad-b180193a0ca7"
      },
      "source": [
        "dataset.shape"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(129920, 9)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vciVZ-N18YwS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6e3f7e76-1691-43a4-934b-ce7debcb1fae"
      },
      "source": [
        "dataset.dtypes"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Date                      object\n",
              "Time                      object\n",
              "Global_active_power       object\n",
              "Global_reactive_power     object\n",
              "Voltage                   object\n",
              "Global_intensity          object\n",
              "Sub_metering_1            object\n",
              "Sub_metering_2            object\n",
              "Sub_metering_3           float64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6C7gQiQp8YwU",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "49d02cf2-c8d0-4851-ca2b-ce976e0a67d8"
      },
      "source": [
        "# Checando se há valores missing\n",
        "dataset.isnull().values.any()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DLghn25R8YwV"
      },
      "source": [
        "# Remove os registros com valores NA e remove as duas primeiras colunas (não são necessárias)\n",
        "dataset = dataset.iloc[0:, 2:9].dropna()"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kNUEgbdc8YwV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "fa5d7a04-a94f-4700-e540-5a580710d010"
      },
      "source": [
        "dataset.head()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Global_active_power</th>\n",
              "      <th>Global_reactive_power</th>\n",
              "      <th>Voltage</th>\n",
              "      <th>Global_intensity</th>\n",
              "      <th>Sub_metering_1</th>\n",
              "      <th>Sub_metering_2</th>\n",
              "      <th>Sub_metering_3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4.216</td>\n",
              "      <td>0.418</td>\n",
              "      <td>234.840</td>\n",
              "      <td>18.400</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5.360</td>\n",
              "      <td>0.436</td>\n",
              "      <td>233.630</td>\n",
              "      <td>23.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>16.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5.374</td>\n",
              "      <td>0.498</td>\n",
              "      <td>233.290</td>\n",
              "      <td>23.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>5.388</td>\n",
              "      <td>0.502</td>\n",
              "      <td>233.740</td>\n",
              "      <td>23.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>3.666</td>\n",
              "      <td>0.528</td>\n",
              "      <td>235.680</td>\n",
              "      <td>15.800</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>17.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Global_active_power Global_reactive_power  ... Sub_metering_2 Sub_metering_3\n",
              "0               4.216                 0.418  ...          1.000           17.0\n",
              "1               5.360                 0.436  ...          1.000           16.0\n",
              "2               5.374                 0.498  ...          2.000           17.0\n",
              "3               5.388                 0.502  ...          1.000           17.0\n",
              "4               3.666                 0.528  ...          1.000           17.0\n",
              "\n",
              "[5 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8aYy0Y5P8YwW",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2ebc41ae-271c-4f3f-d145-cd6593e38287"
      },
      "source": [
        "# Checando se há valores missing\n",
        "dataset.isnull().values.any()"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "False"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5o0M8hi68YwX"
      },
      "source": [
        "# Obtém os valores dos atributos \n",
        "dataset_atrib = dataset.values"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HuYspkMh8YwY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4431f797-f1df-4a75-c526-0072c6dea0fd"
      },
      "source": [
        "dataset_atrib"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([['4.216', '0.418', '234.840', ..., '0.000', '1.000', 17.0],\n",
              "       ['5.360', '0.436', '233.630', ..., '0.000', '1.000', 16.0],\n",
              "       ['5.374', '0.498', '233.290', ..., '0.000', '2.000', 17.0],\n",
              "       ...,\n",
              "       ['2.194', '0.094', '238.700', ..., '0.000', '0.000', 17.0],\n",
              "       ['2.198', '0.094', '238.920', ..., '0.000', '0.000', 18.0],\n",
              "       ['2.208', '0.096', '239.370', ..., '0.000', '0.000', 17.0]],\n",
              "      dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-KxfjyuZ8YwY"
      },
      "source": [
        "# Coleta uma amostra de 1% dos dados para não comprometer a memória do computador\n",
        "amostra1, amostra2 = train_test_split(dataset_atrib, train_size = .01)"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o-itgcYM8YwY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "344adf53-ab61-4151-ddde-5e2edd2fd05f"
      },
      "source": [
        "amostra1.shape"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1299, 7)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GWVr3IJW8YwZ"
      },
      "source": [
        "# Aplica redução de dimensionalidade\n",
        "pca = PCA(n_components = 2).fit_transform(amostra1)"
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fVkrrXts8YwZ"
      },
      "source": [
        "# Determinando um range de K\n",
        "k_range = range(1,12)"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JJ7I199N8YwZ"
      },
      "source": [
        "# Aplicando o modelo K-Means para cada valor de K (esta célula pode levar bastante tempo para ser executada)\n",
        "k_means_var = [KMeans(n_clusters = k).fit(pca) for k in k_range]"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4UXHXhNP8Ywa"
      },
      "source": [
        "# Ajustando o centróide do cluster para cada modelo\n",
        "centroids = [X.cluster_centers_ for X in k_means_var]"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4KLcOyBX8Ywa"
      },
      "source": [
        "https://docs.scipy.org/doc/scipy/reference/generated/scipy.spatial.distance.pdist.html\n",
        "\n",
        "https://docs.scipy.org/doc/scipy/reference/generated/scipy.spatial.distance.cdist.html"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yqgRBAfY8Ywa"
      },
      "source": [
        "# Calculando a distância euclidiana de cada ponto de dado para o centróide\n",
        "k_euclid = [cdist(pca, cent, 'euclidean') for cent in centroids]\n",
        "dist = [np.min(ke, axis = 1) for ke in k_euclid]"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GwIjW82z8Ywb"
      },
      "source": [
        "# Soma dos quadrados das distâncias dentro do cluster\n",
        "soma_quadrados_intra_cluster = [sum(d**2) for d in dist]"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IcCZ0hVF8Ywb"
      },
      "source": [
        "# Soma total dos quadrados\n",
        "soma_total = sum(pdist(pca)**2)/pca.shape[0]"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FLZbqiFd8Ywb"
      },
      "source": [
        "# Soma dos quadrados entre clusters\n",
        "soma_quadrados_inter_cluster = soma_total - soma_quadrados_intra_cluster"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nGXJZFtC8Ywc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "7ddc402e-0518-44ae-b697-9eb15f267424"
      },
      "source": [
        "# Curva de Elbow\n",
        "fig = plt.figure()\n",
        "ax = fig.add_subplot(111)\n",
        "ax.plot(k_range, soma_quadrados_inter_cluster/soma_total * 100, 'b*-')\n",
        "ax.set_ylim((0,100))\n",
        "plt.grid(True)\n",
        "plt.xlabel('Número de Clusters')\n",
        "plt.ylabel('Percentual de Variância Explicada')\n",
        "plt.title('Variância Explicada x Valor de K')"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Variância Explicada x Valor de K')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        },
        {
          "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": "code",
      "metadata": {
        "id": "VvcdalDb8Ywd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4e1fbeaa-fcc5-4602-f1e0-6ba0d0b924f5"
      },
      "source": [
        "# Criando um modelo com K = 8\n",
        "modelo_v1 = KMeans(n_clusters = 8)\n",
        "modelo_v1.fit(pca)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
              "       n_clusters=8, n_init=10, n_jobs=None, precompute_distances='auto',\n",
              "       random_state=None, tol=0.0001, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hFJVnI0D8Ywe"
      },
      "source": [
        "# Obtém os valores mínimos e máximos e organiza o shape\n",
        "x_min, x_max = pca[:, 0].min() - 5, pca[:, 0].max() - 1\n",
        "y_min, y_max = pca[:, 1].min() + 1, pca[:, 1].max() + 5\n",
        "xx, yy = np.meshgrid(np.arange(x_min, x_max, .02), np.arange(y_min, y_max, .02))\n",
        "Z = modelo_v1.predict(np.c_[xx.ravel(), yy.ravel()])\n",
        "Z = Z.reshape(xx.shape)"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y2sLbmdM8Ywe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        },
        "outputId": "745ad0be-6e8f-4071-f403-f0a2087925dc"
      },
      "source": [
        "# Plot das áreas dos clusters\n",
        "plt.figure(1)\n",
        "plt.clf()\n",
        "plt.imshow(Z, \n",
        "           interpolation = 'nearest',\n",
        "           extent = (xx.min(), xx.max(), yy.min(), yy.max()),\n",
        "           cmap = plt.cm.Paired,\n",
        "           aspect = 'auto', \n",
        "           origin = 'lower')"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f390ff75d10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        },
        {
          "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": "code",
      "metadata": {
        "id": "nkYSft_z8Ywg",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "outputId": "3db9d479-e57d-4293-f4e6-6ad92371a65b"
      },
      "source": [
        "# Plot dos centróides\n",
        "plt.plot(pca[:, 0], pca[:, 1], 'k.', markersize = 4)\n",
        "centroids = modelo_v1.cluster_centers_\n",
        "inert = modelo_v1.inertia_\n",
        "plt.scatter(centroids[:, 0], centroids[:, 1], marker = 'x', s = 169, linewidths = 3, color = 'r', zorder = 8)\n",
        "plt.xlim(x_min, x_max)\n",
        "plt.ylim(y_min, y_max)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1UAYYAy38Ywh"
      },
      "source": [
        "?silhouette_score"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u65CL3688Ywi",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6ae58d1e-2cad-4c98-b882-3445e2b73fa7"
      },
      "source": [
        "# Silhouette Score\n",
        "labels = modelo_v1.labels_\n",
        "silhouette_score(pca, labels, metric = 'euclidean')"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6703633017858814"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r5uC1r7F8Ywi",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f371a822-66a1-4aa8-9cb5-bfac8e84e4a8"
      },
      "source": [
        "# Criando um modelo com K = 10\n",
        "modelo_v2 = KMeans(n_clusters = 10)\n",
        "modelo_v2.fit(pca)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
              "       n_clusters=10, n_init=10, n_jobs=None, precompute_distances='auto',\n",
              "       random_state=None, tol=0.0001, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lZy1VR_08Ywj"
      },
      "source": [
        "# Obtém os valores mínimos e máximos e organiza o shape\n",
        "x_min, x_max = pca[:, 0].min() - 5, pca[:, 0].max() - 1\n",
        "y_min, y_max = pca[:, 1].min() + 1, pca[:, 1].max() + 5\n",
        "xx, yy = np.meshgrid(np.arange(x_min, x_max, .02), np.arange(y_min, y_max, .02))\n",
        "Z = modelo_v2.predict(np.c_[xx.ravel(), yy.ravel()])\n",
        "Z = Z.reshape(xx.shape)"
      ],
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HD9LnHKZ8Ywj",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        },
        "outputId": "18a9d7cf-0e43-4e31-b754-34c4f4a502e5"
      },
      "source": [
        "# Plot das áreas dos clusters\n",
        "plt.figure(1)\n",
        "plt.clf()\n",
        "plt.imshow(Z, \n",
        "           interpolation = 'nearest',\n",
        "           extent = (xx.min(), xx.max(), yy.min(), yy.max()),\n",
        "           cmap = plt.cm.Paired,\n",
        "           aspect = 'auto', \n",
        "           origin = 'lower')"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f390ff49fd0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        },
        {
          "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": "code",
      "metadata": {
        "id": "_NSvHmu38Ywj",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "outputId": "33387536-b5af-4a93-91e5-1de91c732b20"
      },
      "source": [
        "# Plot dos centróides\n",
        "plt.plot(pca[:, 0], pca[:, 1], 'k.', markersize = 4)\n",
        "centroids = modelo_v2.cluster_centers_\n",
        "inert = modelo_v2.inertia_\n",
        "plt.scatter(centroids[:, 0], centroids[:, 1], marker = 'x', s = 169, linewidths = 3, color = 'r', zorder = 8)\n",
        "plt.xlim(x_min, x_max)\n",
        "plt.ylim(y_min, y_max)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hW5WzKC68Ywj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e3ea90fd-949b-4cb8-933c-0609f05a0345"
      },
      "source": [
        "# Silhouette Score\n",
        "labels = modelo_v2.labels_\n",
        "silhouette_score(pca, labels, metric = 'euclidean')"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.673264626647019"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rMddz2TM8Ywk"
      },
      "source": [
        "Criando o Cluster Map com os clusters do Modelo V1 que apresentou melhor Silhouette Score."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z8WG1vgj8Ywk"
      },
      "source": [
        "# Lista com nomes das colunas\n",
        "names = ['Global_active_power', 'Global_reactive_power', 'Voltage', 'Global_intensity', 'Sub_metering_1', 'Sub_metering_2', 'Sub_metering_3']"
      ],
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xBr6_T5E8Ywk"
      },
      "source": [
        "# Cria o cluster map\n",
        "cluster_map = pd.DataFrame(amostra1, columns = names)\n",
        "cluster_map['Global_active_power'] = pd.to_numeric(cluster_map['Global_active_power'])\n",
        "cluster_map['cluster'] = modelo_v1.labels_"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9mVXHtaA8Ywl",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "d78f217e-15e3-4cb2-f33a-093242961b79"
      },
      "source": [
        "cluster_map"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Global_active_power</th>\n",
              "      <th>Global_reactive_power</th>\n",
              "      <th>Voltage</th>\n",
              "      <th>Global_intensity</th>\n",
              "      <th>Sub_metering_1</th>\n",
              "      <th>Sub_metering_2</th>\n",
              "      <th>Sub_metering_3</th>\n",
              "      <th>cluster</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3.098</td>\n",
              "      <td>0.218</td>\n",
              "      <td>240.430</td>\n",
              "      <td>13.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>18</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.472</td>\n",
              "      <td>0.362</td>\n",
              "      <td>243.220</td>\n",
              "      <td>2.400</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.208</td>\n",
              "      <td>0.000</td>\n",
              "      <td>243.760</td>\n",
              "      <td>0.800</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.952</td>\n",
              "      <td>0.482</td>\n",
              "      <td>244.940</td>\n",
              "      <td>8.200</td>\n",
              "      <td>2.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.016</td>\n",
              "      <td>0.076</td>\n",
              "      <td>241.470</td>\n",
              "      <td>4.200</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1294</th>\n",
              "      <td>0.332</td>\n",
              "      <td>0.144</td>\n",
              "      <td>240.940</td>\n",
              "      <td>1.600</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1295</th>\n",
              "      <td>0.666</td>\n",
              "      <td>0.122</td>\n",
              "      <td>241.910</td>\n",
              "      <td>4.200</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1296</th>\n",
              "      <td>0.524</td>\n",
              "      <td>0.050</td>\n",
              "      <td>243.570</td>\n",
              "      <td>2.200</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1297</th>\n",
              "      <td>0.972</td>\n",
              "      <td>0.114</td>\n",
              "      <td>242.420</td>\n",
              "      <td>4.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1298</th>\n",
              "      <td>0.230</td>\n",
              "      <td>0.000</td>\n",
              "      <td>242.330</td>\n",
              "      <td>1.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1299 rows × 8 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      Global_active_power Global_reactive_power  ... Sub_metering_3 cluster\n",
              "0                   3.098                 0.218  ...             18       7\n",
              "1                   0.472                 0.362  ...              0       1\n",
              "2                   0.208                 0.000  ...              0       1\n",
              "3                   1.952                 0.482  ...             18       0\n",
              "4                   1.016                 0.076  ...              0       1\n",
              "...                   ...                   ...  ...            ...     ...\n",
              "1294                0.332                 0.144  ...              0       1\n",
              "1295                0.666                 0.122  ...              0       1\n",
              "1296                0.524                 0.050  ...              0       1\n",
              "1297                0.972                 0.114  ...              0       1\n",
              "1298                0.230                 0.000  ...              0       1\n",
              "\n",
              "[1299 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tTH0BzGl8Ywl",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "20eede3b-1abf-43c7-b507-212ececff54e"
      },
      "source": [
        "# Calcula a média de consumo de energia por cluster\n",
        "cluster_map.groupby('cluster')['Global_active_power'].mean()"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "cluster\n",
              "0    1.616276\n",
              "1    0.398535\n",
              "2    4.819800\n",
              "3    4.289355\n",
              "4    2.905000\n",
              "5    8.171333\n",
              "6    2.204556\n",
              "7    3.178194\n",
              "Name: Global_active_power, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nxBxqsO88Ywl"
      },
      "source": [
        "### Fim"
      ]
    }
  ]
}