aula0/Introdução_a_Machine_Learning_Otimização.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Introdução_a_Machine_Learning_Otimização.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "metadata": {
        "id": "y7ik04NlDZMA",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "\n",
        "uri = \"https://gist.githubusercontent.com/guilhermesilveira/e99a526b2e7ccc6c3b70f53db43a87d2/raw/1605fc74aa778066bf2e6695e24d53cf65f2f447/machine-learning-carros-simulacao.csv\"\n",
        "dados = pd.read_csv(uri).drop(columns=[\"Unnamed: 0\"], axis=1)\n",
        "dados.head()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "b52E4e1hjegN",
        "colab_type": "code",
        "outputId": "97420810-da24-4bd7-8b8d-242c06aa220a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "cell_type": "code",
      "source": [
        "# situação horrível de \"azar\" onde as classes estão ordenadas por padrão\n",
        "\n",
        "dados_azar = dados.sort_values(\"vendido\", ascending=True)\n",
        "x_azar = dados_azar[[\"preco\", \"idade_do_modelo\",\"km_por_ano\"]]\n",
        "y_azar = dados_azar[\"vendido\"]\n",
        "dados_azar.head()"
      ],
      "execution_count": 8,
      "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>preco</th>\n",
              "      <th>vendido</th>\n",
              "      <th>idade_do_modelo</th>\n",
              "      <th>km_por_ano</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>4999</th>\n",
              "      <td>74023.29</td>\n",
              "      <td>0</td>\n",
              "      <td>12</td>\n",
              "      <td>24812.80412</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5322</th>\n",
              "      <td>84843.49</td>\n",
              "      <td>0</td>\n",
              "      <td>13</td>\n",
              "      <td>23095.63834</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5319</th>\n",
              "      <td>83100.27</td>\n",
              "      <td>0</td>\n",
              "      <td>19</td>\n",
              "      <td>36240.72746</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5316</th>\n",
              "      <td>87932.13</td>\n",
              "      <td>0</td>\n",
              "      <td>16</td>\n",
              "      <td>32249.56426</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5315</th>\n",
              "      <td>77937.01</td>\n",
              "      <td>0</td>\n",
              "      <td>15</td>\n",
              "      <td>28414.50704</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         preco  vendido  idade_do_modelo   km_por_ano\n",
              "4999  74023.29        0               12  24812.80412\n",
              "5322  84843.49        0               13  23095.63834\n",
              "5319  83100.27        0               19  36240.72746\n",
              "5316  87932.13        0               16  32249.56426\n",
              "5315  77937.01        0               15  28414.50704"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "metadata": {
        "id": "ZNT26vFEeYmz",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import cross_validate\n",
        "from sklearn.dummy import DummyClassifier\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "modelo = DummyClassifier()\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = 10, return_train_score=False)\n",
        "media = results['test_score'].mean()\n",
        "desvio_padrao = results['test_score'].std()\n",
        "print(\"Accuracy com dummy stratified, 10 = [%.2f, %.2f]\" % ((media - 2 * desvio_padrao)*100, (media + 2 * desvio_padrao) * 100))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "73mCcFA_eG_K",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import cross_validate\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "modelo = DecisionTreeClassifier(max_depth=2)\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = 10, return_train_score=False)\n",
        "media = results['test_score'].mean()\n",
        "desvio_padrao = results['test_score'].std()\n",
        "print(\"Accuracy com cross validation, 10 = [%.2f, %.2f]\" % ((media - 2 * desvio_padrao)*100, (media + 2 * desvio_padrao) * 100))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "5C8Y6J-PGpYf",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# gerando dados elatorios de modelo de carro para simulacao de agrupamento ao usar nosso estimador\n",
        "\n",
        "np.random.seed(SEED)\n",
        "dados['modelo'] = dados.idade_do_modelo + np.random.randint(-2, 3, size=10000)\n",
        "dados.modelo = dados.modelo + abs(dados.modelo.min()) + 1\n",
        "dados.head()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "3hmjt7qPHOZY",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def imprime_resultados(results):\n",
        "  media = results['test_score'].mean() * 100\n",
        "  desvio = results['test_score'].std() * 100\n",
        "  print(\"Accuracy médio %.2f\" % media)\n",
        "  print(\"Intervalo [%.2f, %.2f]\" % (media - 2 * desvio, media + 2 * desvio))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "IKD8uaWkNcUL",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# GroupKFold para analisar como o modelo se comporta com novos grupos\n",
        "\n",
        "from sklearn.model_selection import GroupKFold\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "cv = GroupKFold(n_splits = 10)\n",
        "modelo = DecisionTreeClassifier(max_depth=2)\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=False)\n",
        "imprime_resultados(results)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "goijy0rSS7n-",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# GroupKFold em um pipeline com StandardScaler e SVC\n",
        "\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.pipeline import Pipeline\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "scaler = StandardScaler()\n",
        "modelo = SVC()\n",
        "\n",
        "pipeline = Pipeline([('transformacao',scaler), ('estimador',modelo)])\n",
        "\n",
        "cv = GroupKFold(n_splits = 10)\n",
        "results = cross_validate(pipeline, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=False)\n",
        "imprime_resultados(results)"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}