UDP_BA_Clase_08.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jHt4VVsFEnew"
      },
      "source": [
        "![logo_fae_web.jpeg](data:image/jpeg;base64,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)\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yvpM01Wjyn2x"
      },
      "source": [
        "# UDP - Bussines Analytics 1\n",
        "\n",
        "### Contenido de la clase:\n",
        "\n",
        "### Data Processing\n",
        "\n",
        "### Contenido de la clase\n",
        "* Creación de Variables Dummies\n",
        "* One Hot Encoding\n",
        "* Label Encoding\n",
        "* Target Encoding\n",
        "* Ejemplo práctico"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kJij5qsb1OXb"
      },
      "source": [
        "## Hacia donde vamos...\n",
        "* Analítica Descriptiva:\n",
        "  * Preprocessing\n",
        "    * Missing Values\n",
        "    * Outliers\n",
        "    * Feature Engineering\n",
        "    * Feature Selection\n",
        "  * Exploratory Data Analysis\n",
        "    * Análisis Univariado: distribución normal?\n",
        "    * Test de comparación de médias: t-test, Chi2, KS, etc\n",
        "    * Regresión Logística"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kZTtDD0oJDyM"
      },
      "source": [
        "### Instalación de la librería de estadísticas"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 253,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n4zL7ZXF8sUV",
        "outputId": "1c687648-136c-40e8-ddd5-1ad67c6a963c"
      },
      "outputs": [],
      "source": [
        "# Para instalarla debes sacar el signo #\n",
        "# ! pip install scipy"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 254,
      "metadata": {
        "id": "PCQcMbDiEYBr"
      },
      "outputs": [],
      "source": [
        "# Librerías de tratamiento de datos (las principales)\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import missingno as msno\n",
        "\n",
        "\n",
        "# Librerías de gráficas\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "# complementa algunas visualizaciones\n",
        "%matplotlib inline\n",
        "# apply some cool styling\n",
        "sns.set(color_codes=True)\n",
        "\n",
        "# Instalación de librería estadística\n",
        "from scipy import stats\n",
        "from scipy.stats import norm\n",
        "\n",
        "\n",
        "# Librerías de advertencias\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zTSBcqzjhaBi"
      },
      "source": [
        "### Importamos la data desde nuestro Drive\n",
        "\n",
        "1. from google.colab import drive\n",
        "2. drive.mount(\"/content/drive\")\n",
        "3. Permitir acceso a nuestro Google Colab\n",
        "4. link archivo\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 255,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AguH4-H0hhwT",
        "outputId": "6f3ebdda-3ad4-4e9e-d08b-c5150ac45f9c"
      },
      "outputs": [],
      "source": [
        "# from google.colab import drive\n",
        "# drive.mount(\"/content/drive\")\n",
        "# df = pd.read_csv(\"/content/drive/MyDrive/UDP_BA01/data_store.csv\",sep=\",\",encoding=\"latin-1\")\n",
        "\n",
        "# # To display the top 5 rows\n",
        "# df.head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 256,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 585
        },
        "id": "X2zovbrFgQVJ",
        "outputId": "1227a94d-0b3a-4ba8-e0b3-93a75ca0ce09"
      },
      "outputs": [
        {
          "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>Row ID</th>\n",
              "      <th>Order ID</th>\n",
              "      <th>Order Date</th>\n",
              "      <th>Ship Date</th>\n",
              "      <th>Ship Mode</th>\n",
              "      <th>Customer ID</th>\n",
              "      <th>Customer Name</th>\n",
              "      <th>Segment</th>\n",
              "      <th>Country</th>\n",
              "      <th>City</th>\n",
              "      <th>...</th>\n",
              "      <th>Postal Code</th>\n",
              "      <th>Region</th>\n",
              "      <th>Product ID</th>\n",
              "      <th>Category</th>\n",
              "      <th>Sub-Category</th>\n",
              "      <th>Product Name</th>\n",
              "      <th>Sales</th>\n",
              "      <th>Quantity</th>\n",
              "      <th>Discount</th>\n",
              "      <th>Profit</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>3017</th>\n",
              "      <td>3018</td>\n",
              "      <td>US-2016-160528</td>\n",
              "      <td>8/23/2016</td>\n",
              "      <td>8/30/2016</td>\n",
              "      <td>Standard Class</td>\n",
              "      <td>MH-18115</td>\n",
              "      <td>Mick Hernandez</td>\n",
              "      <td>Home Office</td>\n",
              "      <td>United States</td>\n",
              "      <td>Pharr</td>\n",
              "      <td>...</td>\n",
              "      <td>78577</td>\n",
              "      <td>Central</td>\n",
              "      <td>FUR-FU-10004973</td>\n",
              "      <td>Furniture</td>\n",
              "      <td>Furnishings</td>\n",
              "      <td>Flat Face Poster Frame</td>\n",
              "      <td>22.608</td>\n",
              "      <td>3</td>\n",
              "      <td>0.6</td>\n",
              "      <td>-10.1736</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5053</th>\n",
              "      <td>5054</td>\n",
              "      <td>CA-2015-141243</td>\n",
              "      <td>1/3/2015</td>\n",
              "      <td>1/8/2015</td>\n",
              "      <td>Second Class</td>\n",
              "      <td>AH-10465</td>\n",
              "      <td>Amy Hunt</td>\n",
              "      <td>Consumer</td>\n",
              "      <td>United States</td>\n",
              "      <td>Dallas</td>\n",
              "      <td>...</td>\n",
              "      <td>75217</td>\n",
              "      <td>Central</td>\n",
              "      <td>TEC-AC-10003198</td>\n",
              "      <td>Technology</td>\n",
              "      <td>Accessories</td>\n",
              "      <td>Enermax Acrylux Wireless Keyboard</td>\n",
              "      <td>398.400</td>\n",
              "      <td>5</td>\n",
              "      <td>0.2</td>\n",
              "      <td>84.6600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8003</th>\n",
              "      <td>8004</td>\n",
              "      <td>CA-2014-143210</td>\n",
              "      <td>12/1/2014</td>\n",
              "      <td>12/3/2014</td>\n",
              "      <td>First Class</td>\n",
              "      <td>AA-10645</td>\n",
              "      <td>Anna Andreadi</td>\n",
              "      <td>Consumer</td>\n",
              "      <td>United States</td>\n",
              "      <td>Lowell</td>\n",
              "      <td>...</td>\n",
              "      <td>1852</td>\n",
              "      <td>East</td>\n",
              "      <td>TEC-PH-10004434</td>\n",
              "      <td>Technology</td>\n",
              "      <td>Phones</td>\n",
              "      <td>Cisco IP Phone 7961G VoIP phone - Dark gray</td>\n",
              "      <td>271.900</td>\n",
              "      <td>2</td>\n",
              "      <td>0.0</td>\n",
              "      <td>78.8510</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6556</th>\n",
              "      <td>6557</td>\n",
              "      <td>CA-2014-137092</td>\n",
              "      <td>10/20/2014</td>\n",
              "      <td>10/22/2014</td>\n",
              "      <td>Second Class</td>\n",
              "      <td>LS-16975</td>\n",
              "      <td>Lindsay Shagiari</td>\n",
              "      <td>Home Office</td>\n",
              "      <td>United States</td>\n",
              "      <td>Chicago</td>\n",
              "      <td>...</td>\n",
              "      <td>60653</td>\n",
              "      <td>Central</td>\n",
              "      <td>OFF-BI-10000632</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Binders</td>\n",
              "      <td>Satellite Sectional Post Binders</td>\n",
              "      <td>8.682</td>\n",
              "      <td>1</td>\n",
              "      <td>0.8</td>\n",
              "      <td>-14.7594</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5571</th>\n",
              "      <td>5572</td>\n",
              "      <td>CA-2014-151162</td>\n",
              "      <td>12/5/2014</td>\n",
              "      <td>12/9/2014</td>\n",
              "      <td>Standard Class</td>\n",
              "      <td>EA-14035</td>\n",
              "      <td>Erin Ashbrook</td>\n",
              "      <td>Corporate</td>\n",
              "      <td>United States</td>\n",
              "      <td>Columbus</td>\n",
              "      <td>...</td>\n",
              "      <td>43229</td>\n",
              "      <td>East</td>\n",
              "      <td>OFF-PA-10001033</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Paper</td>\n",
              "      <td>Xerox 1893</td>\n",
              "      <td>98.376</td>\n",
              "      <td>3</td>\n",
              "      <td>0.2</td>\n",
              "      <td>35.6613</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6997</th>\n",
              "      <td>6998</td>\n",
              "      <td>CA-2017-117443</td>\n",
              "      <td>12/23/2017</td>\n",
              "      <td>12/25/2017</td>\n",
              "      <td>Second Class</td>\n",
              "      <td>JB-15400</td>\n",
              "      <td>Jennifer Braxton</td>\n",
              "      <td>Corporate</td>\n",
              "      <td>United States</td>\n",
              "      <td>Rockford</td>\n",
              "      <td>...</td>\n",
              "      <td>61107</td>\n",
              "      <td>Central</td>\n",
              "      <td>OFF-PA-10004475</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Paper</td>\n",
              "      <td>Xerox 1940</td>\n",
              "      <td>175.872</td>\n",
              "      <td>4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>63.7536</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7420</th>\n",
              "      <td>7421</td>\n",
              "      <td>US-2016-114013</td>\n",
              "      <td>3/13/2016</td>\n",
              "      <td>3/15/2016</td>\n",
              "      <td>Second Class</td>\n",
              "      <td>SC-20770</td>\n",
              "      <td>Stewart Carmichael</td>\n",
              "      <td>Corporate</td>\n",
              "      <td>United States</td>\n",
              "      <td>Philadelphia</td>\n",
              "      <td>...</td>\n",
              "      <td>19134</td>\n",
              "      <td>East</td>\n",
              "      <td>OFF-BI-10002353</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Binders</td>\n",
              "      <td>GBC VeloBind Cover Sets</td>\n",
              "      <td>23.160</td>\n",
              "      <td>5</td>\n",
              "      <td>0.7</td>\n",
              "      <td>-15.4400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8511</th>\n",
              "      <td>8512</td>\n",
              "      <td>CA-2016-103359</td>\n",
              "      <td>12/22/2016</td>\n",
              "      <td>12/26/2016</td>\n",
              "      <td>Standard Class</td>\n",
              "      <td>LC-16885</td>\n",
              "      <td>Lena Creighton</td>\n",
              "      <td>Consumer</td>\n",
              "      <td>United States</td>\n",
              "      <td>New York City</td>\n",
              "      <td>...</td>\n",
              "      <td>10035</td>\n",
              "      <td>East</td>\n",
              "      <td>OFF-BI-10004233</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Binders</td>\n",
              "      <td>GBC Pre-Punched Binding Paper, Plastic, White,...</td>\n",
              "      <td>25.584</td>\n",
              "      <td>2</td>\n",
              "      <td>0.2</td>\n",
              "      <td>8.9544</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8425</th>\n",
              "      <td>8426</td>\n",
              "      <td>CA-2016-137652</td>\n",
              "      <td>8/28/2016</td>\n",
              "      <td>8/30/2016</td>\n",
              "      <td>First Class</td>\n",
              "      <td>EB-13750</td>\n",
              "      <td>Edward Becker</td>\n",
              "      <td>Corporate</td>\n",
              "      <td>United States</td>\n",
              "      <td>Cincinnati</td>\n",
              "      <td>...</td>\n",
              "      <td>45231</td>\n",
              "      <td>East</td>\n",
              "      <td>OFF-BI-10004099</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Binders</td>\n",
              "      <td>GBC VeloBinder Strips</td>\n",
              "      <td>18.432</td>\n",
              "      <td>8</td>\n",
              "      <td>0.7</td>\n",
              "      <td>-12.2880</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7683</th>\n",
              "      <td>7684</td>\n",
              "      <td>CA-2015-120782</td>\n",
              "      <td>4/28/2015</td>\n",
              "      <td>5/1/2015</td>\n",
              "      <td>First Class</td>\n",
              "      <td>SD-20485</td>\n",
              "      <td>Shirley Daniels</td>\n",
              "      <td>Home Office</td>\n",
              "      <td>United States</td>\n",
              "      <td>Midland</td>\n",
              "      <td>...</td>\n",
              "      <td>48640</td>\n",
              "      <td>Central</td>\n",
              "      <td>OFF-BI-10003527</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Binders</td>\n",
              "      <td>Fellowes PB500 Electric Punch Plastic Comb Bin...</td>\n",
              "      <td>3812.970</td>\n",
              "      <td>3</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1906.4850</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>10 rows × 21 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      Row ID        Order ID  Order Date   Ship Date       Ship Mode  \\\n",
              "3017    3018  US-2016-160528   8/23/2016   8/30/2016  Standard Class   \n",
              "5053    5054  CA-2015-141243    1/3/2015    1/8/2015    Second Class   \n",
              "8003    8004  CA-2014-143210   12/1/2014   12/3/2014     First Class   \n",
              "6556    6557  CA-2014-137092  10/20/2014  10/22/2014    Second Class   \n",
              "5571    5572  CA-2014-151162   12/5/2014   12/9/2014  Standard Class   \n",
              "6997    6998  CA-2017-117443  12/23/2017  12/25/2017    Second Class   \n",
              "7420    7421  US-2016-114013   3/13/2016   3/15/2016    Second Class   \n",
              "8511    8512  CA-2016-103359  12/22/2016  12/26/2016  Standard Class   \n",
              "8425    8426  CA-2016-137652   8/28/2016   8/30/2016     First Class   \n",
              "7683    7684  CA-2015-120782   4/28/2015    5/1/2015     First Class   \n",
              "\n",
              "     Customer ID       Customer Name      Segment        Country  \\\n",
              "3017    MH-18115      Mick Hernandez  Home Office  United States   \n",
              "5053    AH-10465            Amy Hunt     Consumer  United States   \n",
              "8003    AA-10645       Anna Andreadi     Consumer  United States   \n",
              "6556    LS-16975    Lindsay Shagiari  Home Office  United States   \n",
              "5571    EA-14035       Erin Ashbrook    Corporate  United States   \n",
              "6997    JB-15400    Jennifer Braxton    Corporate  United States   \n",
              "7420    SC-20770  Stewart Carmichael    Corporate  United States   \n",
              "8511    LC-16885      Lena Creighton     Consumer  United States   \n",
              "8425    EB-13750       Edward Becker    Corporate  United States   \n",
              "7683    SD-20485     Shirley Daniels  Home Office  United States   \n",
              "\n",
              "               City  ... Postal Code   Region       Product ID  \\\n",
              "3017          Pharr  ...       78577  Central  FUR-FU-10004973   \n",
              "5053         Dallas  ...       75217  Central  TEC-AC-10003198   \n",
              "8003         Lowell  ...        1852     East  TEC-PH-10004434   \n",
              "6556        Chicago  ...       60653  Central  OFF-BI-10000632   \n",
              "5571       Columbus  ...       43229     East  OFF-PA-10001033   \n",
              "6997       Rockford  ...       61107  Central  OFF-PA-10004475   \n",
              "7420   Philadelphia  ...       19134     East  OFF-BI-10002353   \n",
              "8511  New York City  ...       10035     East  OFF-BI-10004233   \n",
              "8425     Cincinnati  ...       45231     East  OFF-BI-10004099   \n",
              "7683        Midland  ...       48640  Central  OFF-BI-10003527   \n",
              "\n",
              "             Category Sub-Category  \\\n",
              "3017        Furniture  Furnishings   \n",
              "5053       Technology  Accessories   \n",
              "8003       Technology       Phones   \n",
              "6556  Office Supplies      Binders   \n",
              "5571  Office Supplies        Paper   \n",
              "6997  Office Supplies        Paper   \n",
              "7420  Office Supplies      Binders   \n",
              "8511  Office Supplies      Binders   \n",
              "8425  Office Supplies      Binders   \n",
              "7683  Office Supplies      Binders   \n",
              "\n",
              "                                           Product Name     Sales  Quantity  \\\n",
              "3017                             Flat Face Poster Frame    22.608         3   \n",
              "5053                  Enermax Acrylux Wireless Keyboard   398.400         5   \n",
              "8003        Cisco IP Phone 7961G VoIP phone - Dark gray   271.900         2   \n",
              "6556                   Satellite Sectional Post Binders     8.682         1   \n",
              "5571                                         Xerox 1893    98.376         3   \n",
              "6997                                         Xerox 1940   175.872         4   \n",
              "7420                            GBC VeloBind Cover Sets    23.160         5   \n",
              "8511  GBC Pre-Punched Binding Paper, Plastic, White,...    25.584         2   \n",
              "8425                              GBC VeloBinder Strips    18.432         8   \n",
              "7683  Fellowes PB500 Electric Punch Plastic Comb Bin...  3812.970         3   \n",
              "\n",
              "      Discount     Profit  \n",
              "3017       0.6   -10.1736  \n",
              "5053       0.2    84.6600  \n",
              "8003       0.0    78.8510  \n",
              "6556       0.8   -14.7594  \n",
              "5571       0.2    35.6613  \n",
              "6997       0.2    63.7536  \n",
              "7420       0.7   -15.4400  \n",
              "8511       0.2     8.9544  \n",
              "8425       0.7   -12.2880  \n",
              "7683       0.0  1906.4850  \n",
              "\n",
              "[10 rows x 21 columns]"
            ]
          },
          "execution_count": 256,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Data Local\n",
        "df = pd.read_csv(\"/Users/franciscoletelier/Dropbox (Personal)/UDP/UDP_BA/data_store.csv\",sep=\",\",encoding=\"latin-1\")\n",
        "df.sample(10)\n",
        "\n",
        "\n",
        "\n",
        "# Data desde Github\n",
        "# df = pd.read_csv(\"data_store.csv\",sep=\",\",encoding=\"latin-1\")\n",
        "# df.sample(10)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "98qjC_KTsTqi"
      },
      "source": [
        "# Prepocesamiento de la data\n",
        "## Variable Objetivo: Profit"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### Eliminación de columnas innecesarías"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 257,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 293
        },
        "id": "IMOgGLPBgQOX",
        "outputId": "048f735b-1c13-4e37-8fb2-e5a6a6ce9e0f"
      },
      "outputs": [
        {
          "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>Row ID</th>\n",
              "      <th>Order Date</th>\n",
              "      <th>Ship Date</th>\n",
              "      <th>Ship Mode</th>\n",
              "      <th>Segment</th>\n",
              "      <th>City</th>\n",
              "      <th>State</th>\n",
              "      <th>Region</th>\n",
              "      <th>Category</th>\n",
              "      <th>Sub-Category</th>\n",
              "      <th>Product Name</th>\n",
              "      <th>Sales</th>\n",
              "      <th>Quantity</th>\n",
              "      <th>Discount</th>\n",
              "      <th>Profit</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>11/8/2016</td>\n",
              "      <td>11/11/2016</td>\n",
              "      <td>Second Class</td>\n",
              "      <td>Consumer</td>\n",
              "      <td>Henderson</td>\n",
              "      <td>Kentucky</td>\n",
              "      <td>South</td>\n",
              "      <td>Furniture</td>\n",
              "      <td>Bookcases</td>\n",
              "      <td>Bush Somerset Collection Bookcase</td>\n",
              "      <td>261.9600</td>\n",
              "      <td>2</td>\n",
              "      <td>0.00</td>\n",
              "      <td>41.9136</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>11/8/2016</td>\n",
              "      <td>11/11/2016</td>\n",
              "      <td>Second Class</td>\n",
              "      <td>Consumer</td>\n",
              "      <td>Henderson</td>\n",
              "      <td>Kentucky</td>\n",
              "      <td>South</td>\n",
              "      <td>Furniture</td>\n",
              "      <td>Chairs</td>\n",
              "      <td>Hon Deluxe Fabric Upholstered Stacking Chairs,...</td>\n",
              "      <td>731.9400</td>\n",
              "      <td>3</td>\n",
              "      <td>0.00</td>\n",
              "      <td>219.5820</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>6/12/2016</td>\n",
              "      <td>6/16/2016</td>\n",
              "      <td>Second Class</td>\n",
              "      <td>Corporate</td>\n",
              "      <td>Los Angeles</td>\n",
              "      <td>California</td>\n",
              "      <td>West</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Labels</td>\n",
              "      <td>Self-Adhesive Address Labels for Typewriters b...</td>\n",
              "      <td>14.6200</td>\n",
              "      <td>2</td>\n",
              "      <td>0.00</td>\n",
              "      <td>6.8714</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>10/11/2015</td>\n",
              "      <td>10/18/2015</td>\n",
              "      <td>Standard Class</td>\n",
              "      <td>Consumer</td>\n",
              "      <td>Fort Lauderdale</td>\n",
              "      <td>Florida</td>\n",
              "      <td>South</td>\n",
              "      <td>Furniture</td>\n",
              "      <td>Tables</td>\n",
              "      <td>Bretford CR4500 Series Slim Rectangular Table</td>\n",
              "      <td>957.5775</td>\n",
              "      <td>5</td>\n",
              "      <td>0.45</td>\n",
              "      <td>-383.0310</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>10/11/2015</td>\n",
              "      <td>10/18/2015</td>\n",
              "      <td>Standard Class</td>\n",
              "      <td>Consumer</td>\n",
              "      <td>Fort Lauderdale</td>\n",
              "      <td>Florida</td>\n",
              "      <td>South</td>\n",
              "      <td>Office Supplies</td>\n",
              "      <td>Storage</td>\n",
              "      <td>Eldon Fold 'N Roll Cart System</td>\n",
              "      <td>22.3680</td>\n",
              "      <td>2</td>\n",
              "      <td>0.20</td>\n",
              "      <td>2.5164</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Row ID  Order Date   Ship Date       Ship Mode    Segment             City  \\\n",
              "0       1   11/8/2016  11/11/2016    Second Class   Consumer        Henderson   \n",
              "1       2   11/8/2016  11/11/2016    Second Class   Consumer        Henderson   \n",
              "2       3   6/12/2016   6/16/2016    Second Class  Corporate      Los Angeles   \n",
              "3       4  10/11/2015  10/18/2015  Standard Class   Consumer  Fort Lauderdale   \n",
              "4       5  10/11/2015  10/18/2015  Standard Class   Consumer  Fort Lauderdale   \n",
              "\n",
              "        State Region         Category Sub-Category  \\\n",
              "0    Kentucky  South        Furniture    Bookcases   \n",
              "1    Kentucky  South        Furniture       Chairs   \n",
              "2  California   West  Office Supplies       Labels   \n",
              "3     Florida  South        Furniture       Tables   \n",
              "4     Florida  South  Office Supplies      Storage   \n",
              "\n",
              "                                        Product Name     Sales  Quantity  \\\n",
              "0                  Bush Somerset Collection Bookcase  261.9600         2   \n",
              "1  Hon Deluxe Fabric Upholstered Stacking Chairs,...  731.9400         3   \n",
              "2  Self-Adhesive Address Labels for Typewriters b...   14.6200         2   \n",
              "3      Bretford CR4500 Series Slim Rectangular Table  957.5775         5   \n",
              "4                     Eldon Fold 'N Roll Cart System   22.3680         2   \n",
              "\n",
              "   Discount    Profit  \n",
              "0      0.00   41.9136  \n",
              "1      0.00  219.5820  \n",
              "2      0.00    6.8714  \n",
              "3      0.45 -383.0310  \n",
              "4      0.20    2.5164  "
            ]
          },
          "execution_count": 257,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se eliminan las columnas que no se usarán (drop)\n",
        "df = df.drop(['Customer ID','Customer Name', 'Order ID', 'Country', 'Postal Code', 'Product ID'],axis=1)\n",
        "df.head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 258,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 768
        },
        "id": "wWqVBimagQJK",
        "outputId": "4e67c107-d073-4376-dc05-6f28ea8403d9"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<AxesSubplot: >"
            ]
          },
          "execution_count": 258,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1800x720 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "msno.matrix(df)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Creación de Variables Dummy\n",
        "\n",
        "### Creación de variables dummy para la variable categorica: Ship Mode"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 259,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Standard Class    5968\n",
              "Second Class      1945\n",
              "First Class       1538\n",
              "Same Day           543\n",
              "Name: Ship Mode, dtype: int64"
            ]
          },
          "execution_count": 259,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df['Ship Mode'].value_counts()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 260,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Simplificamos las respuestas de la columna Ship Mode\n",
        "df['Ship Mode'] = df['Ship Mode'].replace(['Standard Class','Second Class','First Class','Same Day'],['Standard','Second','First','Express'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 261,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Consumer       5191\n",
              "Corporate      3020\n",
              "Home Office    1783\n",
              "Name: Segment, dtype: int64"
            ]
          },
          "execution_count": 261,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df['Segment'].value_counts()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 262,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Simplificamos las respuestas de la columna Segment\n",
        "df['Segment'] = df['Segment'].replace(['Consumer','Corporate','Home Office'],['Consumidor','Corporativo','Oficina'])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### Creamun subset de trabajo para los ejemplos"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 263,
      "metadata": {},
      "outputs": [
        {
          "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>Ship Mode</th>\n",
              "      <th>Segment</th>\n",
              "      <th>Profit</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Second</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>41.9136</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Second</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>219.5820</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Second</td>\n",
              "      <td>Corporativo</td>\n",
              "      <td>6.8714</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Standard</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>-383.0310</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Standard</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>2.5164</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Ship Mode      Segment    Profit\n",
              "0    Second   Consumidor   41.9136\n",
              "1    Second   Consumidor  219.5820\n",
              "2    Second  Corporativo    6.8714\n",
              "3  Standard   Consumidor -383.0310\n",
              "4  Standard   Consumidor    2.5164"
            ]
          },
          "execution_count": 263,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "subset = df[['Ship Mode','Segment','Profit']]\n",
        "subset.head(5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### One Hot Encoding\n",
        "\n",
        "* One Hot Encoding es un proceso mediante el cual se crea una variable dummy por cada categoría de la variable categórica.\n",
        "* Es decir, si tenemos una variable categórica con 3 categorías, se crearán 3 variables dummy.\n",
        "* La variable dummy toma el valor de 1 si la observación pertenece a esa categoría, y 0 si no pertenece a esa categoría.\n",
        "* Es importante eliminar una de las variables dummy creadas, ya que si no se hace, se puede caer en el problema de multicolinealidad.\n",
        "* Consultar la librería de pandas para crear variables dummy: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.get_dummies.html\n",
        "* Consultar la librería de sklearn para crear variables dummy: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 264,
      "metadata": {},
      "outputs": [
        {
          "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>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Express  First  Second  Standard\n",
              "0        0      0       1         0\n",
              "1        0      0       1         0\n",
              "2        0      0       1         0\n",
              "3        0      0       0         1\n",
              "4        0      0       0         1"
            ]
          },
          "execution_count": 264,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Desarrollaremos el ejemplo con la columna Ship Mode con la librería Pandas: pandas.get_dummies()\n",
        "\n",
        "# Se crea una nueva columna para cada valor de la columna Ship Mode\n",
        "tmp_dummy = pd.get_dummies(subset['Ship Mode'])\n",
        "\n",
        "tmp_dummy.head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 265,
      "metadata": {},
      "outputs": [
        {
          "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>Ship Mode</th>\n",
              "      <th>Segment</th>\n",
              "      <th>Profit</th>\n",
              "      <th>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Second</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>41.9136</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Second</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>219.5820</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Second</td>\n",
              "      <td>Corporativo</td>\n",
              "      <td>6.8714</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Standard</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>-383.0310</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Standard</td>\n",
              "      <td>Consumidor</td>\n",
              "      <td>2.5164</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Ship Mode      Segment    Profit  Express  First  Second  Standard\n",
              "0    Second   Consumidor   41.9136        0      0       1         0\n",
              "1    Second   Consumidor  219.5820        0      0       1         0\n",
              "2    Second  Corporativo    6.8714        0      0       1         0\n",
              "3  Standard   Consumidor -383.0310        0      0       0         1\n",
              "4  Standard   Consumidor    2.5164        0      0       0         1"
            ]
          },
          "execution_count": 265,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se agrega la columna Ship Mode al dataframe\n",
        "subset = pd.concat([subset,tmp_dummy],axis=1)\n",
        "\n",
        "subset.head(5)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### Para evitar multicolinealidad, debemos dropear la columna de referencia"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 266,
      "metadata": {},
      "outputs": [
        {
          "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>Segment</th>\n",
              "      <th>Profit</th>\n",
              "      <th>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>41.9136</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>219.5820</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Corporativo</td>\n",
              "      <td>6.8714</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>-383.0310</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>2.5164</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       Segment    Profit  Express  First  Second  Standard\n",
              "0   Consumidor   41.9136        0      0       1         0\n",
              "1   Consumidor  219.5820        0      0       1         0\n",
              "2  Corporativo    6.8714        0      0       1         0\n",
              "3   Consumidor -383.0310        0      0       0         1\n",
              "4   Consumidor    2.5164        0      0       0         1"
            ]
          },
          "execution_count": 266,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Eliminamos la columna Ship Mode\n",
        "subset = subset.drop(['Ship Mode'],axis=1)\n",
        "subset.head(5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### Desarrollaremos el ejemplo con la columna Segment con la librería Sklearn: sklearn.preprocessing.OneHotEncoder()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 267,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Se importa la librería\n",
        "from sklearn.preprocessing import OneHotEncoder"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 268,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array(['Consumidor', 'Corporativo', 'Oficina'], dtype=object)"
            ]
          },
          "execution_count": 268,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "subset.Segment.unique()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 269,
      "metadata": {},
      "outputs": [
        {
          "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>Consumidor</th>\n",
              "      <th>Corporativo</th>\n",
              "      <th>Oficina</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Consumidor  Corporativo  Oficina\n",
              "0         1.0          0.0      0.0\n",
              "1         1.0          0.0      0.0\n",
              "2         0.0          1.0      0.0\n",
              "3         1.0          0.0      0.0\n",
              "4         1.0          0.0      0.0"
            ]
          },
          "execution_count": 269,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se crea el objeto\n",
        "ohe = OneHotEncoder()\n",
        "\n",
        "# Se aplica el objeto a la columna Segment\n",
        "tmp_ohe = pd.DataFrame(ohe.fit_transform(subset[['Segment']]).toarray()) # .astype(int)\n",
        "\n",
        "# Se agrega el nombre de las columnas\n",
        "tmp_ohe.columns = ['Consumidor', 'Corporativo', 'Oficina']\n",
        "\n",
        "tmp_ohe.head(5)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 270,
      "metadata": {},
      "outputs": [
        {
          "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>Segment</th>\n",
              "      <th>Profit</th>\n",
              "      <th>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "      <th>Consumidor</th>\n",
              "      <th>Corporativo</th>\n",
              "      <th>Oficina</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>41.9136</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>219.5820</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Corporativo</td>\n",
              "      <td>6.8714</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>-383.0310</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Consumidor</td>\n",
              "      <td>2.5164</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       Segment    Profit  Express  First  Second  Standard  Consumidor  \\\n",
              "0   Consumidor   41.9136        0      0       1         0         1.0   \n",
              "1   Consumidor  219.5820        0      0       1         0         1.0   \n",
              "2  Corporativo    6.8714        0      0       1         0         0.0   \n",
              "3   Consumidor -383.0310        0      0       0         1         1.0   \n",
              "4   Consumidor    2.5164        0      0       0         1         1.0   \n",
              "\n",
              "   Corporativo  Oficina  \n",
              "0          0.0      0.0  \n",
              "1          0.0      0.0  \n",
              "2          1.0      0.0  \n",
              "3          0.0      0.0  \n",
              "4          0.0      0.0  "
            ]
          },
          "execution_count": 270,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se agrega la columna tmp_ohe al dataframe\n",
        "subset = pd.concat([subset,tmp_ohe],axis=1)\n",
        "subset.head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 271,
      "metadata": {},
      "outputs": [
        {
          "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>Profit</th>\n",
              "      <th>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "      <th>Consumidor</th>\n",
              "      <th>Corporativo</th>\n",
              "      <th>Oficina</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>41.9136</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>219.5820</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>6.8714</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-383.0310</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2.5164</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     Profit  Express  First  Second  Standard  Consumidor  Corporativo  \\\n",
              "0   41.9136        0      0       1         0         1.0          0.0   \n",
              "1  219.5820        0      0       1         0         1.0          0.0   \n",
              "2    6.8714        0      0       1         0         0.0          1.0   \n",
              "3 -383.0310        0      0       0         1         1.0          0.0   \n",
              "4    2.5164        0      0       0         1         1.0          0.0   \n",
              "\n",
              "   Oficina  \n",
              "0      0.0  \n",
              "1      0.0  \n",
              "2      0.0  \n",
              "3      0.0  \n",
              "4      0.0  "
            ]
          },
          "execution_count": 271,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se eliminan las columnas que no se usarán (drop)\n",
        "subset = subset.drop(['Segment'],axis=1)\n",
        "subset.head(5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Crearemos una variable categórica para clasificar la variable Profit en 2 categorías\n",
        "* Profit > 0: 1\n",
        "* Profit <= 0: 0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 272,
      "metadata": {},
      "outputs": [
        {
          "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>Profit</th>\n",
              "      <th>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "      <th>Consumidor</th>\n",
              "      <th>Corporativo</th>\n",
              "      <th>Oficina</th>\n",
              "      <th>Profit_cat</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>41.9136</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>219.5820</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>6.8714</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-383.0310</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Perdida</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2.5164</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     Profit  Express  First  Second  Standard  Consumidor  Corporativo  \\\n",
              "0   41.9136        0      0       1         0         1.0          0.0   \n",
              "1  219.5820        0      0       1         0         1.0          0.0   \n",
              "2    6.8714        0      0       1         0         0.0          1.0   \n",
              "3 -383.0310        0      0       0         1         1.0          0.0   \n",
              "4    2.5164        0      0       0         1         1.0          0.0   \n",
              "\n",
              "   Oficina Profit_cat  \n",
              "0      0.0   Ganancia  \n",
              "1      0.0   Ganancia  \n",
              "2      0.0   Ganancia  \n",
              "3      0.0    Perdida  \n",
              "4      0.0   Ganancia  "
            ]
          },
          "execution_count": 272,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se crea la variable categorica para la columna Profit con las siguientes condiciones\n",
        "# Profit > 0 : Ganancia\n",
        "# Profit <= 0 : Perdida\n",
        "\n",
        "subset['Profit_cat'] = np.where(subset['Profit'] > 0, 'Ganancia', 'Perdida')\n",
        "subset.head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 273,
      "metadata": {},
      "outputs": [
        {
          "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>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "      <th>Consumidor</th>\n",
              "      <th>Corporativo</th>\n",
              "      <th>Oficina</th>\n",
              "      <th>Profit_cat</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Perdida</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>Ganancia</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Express  First  Second  Standard  Consumidor  Corporativo  Oficina  \\\n",
              "0        0      0       1         0         1.0          0.0      0.0   \n",
              "1        0      0       1         0         1.0          0.0      0.0   \n",
              "2        0      0       1         0         0.0          1.0      0.0   \n",
              "3        0      0       0         1         1.0          0.0      0.0   \n",
              "4        0      0       0         1         1.0          0.0      0.0   \n",
              "\n",
              "  Profit_cat  \n",
              "0   Ganancia  \n",
              "1   Ganancia  \n",
              "2   Ganancia  \n",
              "3    Perdida  \n",
              "4   Ganancia  "
            ]
          },
          "execution_count": 273,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se dropea la columna Profit para no tenerla duplicada\n",
        "subset = subset.drop(['Profit'],axis=1)\n",
        "subset.head(5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Label Encoding\n",
        "* Label Encoding es un proceso mediante el cual se asigna un número a cada categoría de la variable categórica.\n",
        "* Es decir, si tenemos una variable categórica con 3 categorías, se asignarán los números 1, 2 y 3 a cada categoría.\n",
        "* Es importante tener en cuenta que no se debe usar Label Encoding cuando se trabaja con algoritmos de Machine Learning basados en distancias, como KNN o K-Means.\n",
        "* Consultar la librería de pandas para crear variables dummy: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.get_dummies.html\n",
        "* Consultar la librería de sklearn para crear variables dummy: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 274,
      "metadata": {},
      "outputs": [
        {
          "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>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "      <th>Consumidor</th>\n",
              "      <th>Corporativo</th>\n",
              "      <th>Oficina</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>9994.000000</td>\n",
              "      <td>9994.000000</td>\n",
              "      <td>9994.000000</td>\n",
              "      <td>9994.000000</td>\n",
              "      <td>9994.000000</td>\n",
              "      <td>9994.000000</td>\n",
              "      <td>9994.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.054333</td>\n",
              "      <td>0.153892</td>\n",
              "      <td>0.194617</td>\n",
              "      <td>0.597158</td>\n",
              "      <td>0.519412</td>\n",
              "      <td>0.302181</td>\n",
              "      <td>0.178407</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.226684</td>\n",
              "      <td>0.360864</td>\n",
              "      <td>0.395925</td>\n",
              "      <td>0.490494</td>\n",
              "      <td>0.499648</td>\n",
              "      <td>0.459226</td>\n",
              "      <td>0.382874</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "           Express        First       Second     Standard   Consumidor  \\\n",
              "count  9994.000000  9994.000000  9994.000000  9994.000000  9994.000000   \n",
              "mean      0.054333     0.153892     0.194617     0.597158     0.519412   \n",
              "std       0.226684     0.360864     0.395925     0.490494     0.499648   \n",
              "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
              "25%       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
              "50%       0.000000     0.000000     0.000000     1.000000     1.000000   \n",
              "75%       0.000000     0.000000     0.000000     1.000000     1.000000   \n",
              "max       1.000000     1.000000     1.000000     1.000000     1.000000   \n",
              "\n",
              "       Corporativo      Oficina  \n",
              "count  9994.000000  9994.000000  \n",
              "mean      0.302181     0.178407  \n",
              "std       0.459226     0.382874  \n",
              "min       0.000000     0.000000  \n",
              "25%       0.000000     0.000000  \n",
              "50%       0.000000     0.000000  \n",
              "75%       1.000000     0.000000  \n",
              "max       1.000000     1.000000  "
            ]
          },
          "execution_count": 274,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "subset.describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 275,
      "metadata": {},
      "outputs": [
        {
          "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>Express</th>\n",
              "      <th>First</th>\n",
              "      <th>Second</th>\n",
              "      <th>Standard</th>\n",
              "      <th>Consumidor</th>\n",
              "      <th>Corporativo</th>\n",
              "      <th>Oficina</th>\n",
              "      <th>Profit_cat</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Express  First  Second  Standard  Consumidor  Corporativo  Oficina  \\\n",
              "0        0      0       1         0         1.0          0.0      0.0   \n",
              "1        0      0       1         0         1.0          0.0      0.0   \n",
              "2        0      0       1         0         0.0          1.0      0.0   \n",
              "3        0      0       0         1         1.0          0.0      0.0   \n",
              "4        0      0       0         1         1.0          0.0      0.0   \n",
              "\n",
              "   Profit_cat  \n",
              "0           0  \n",
              "1           0  \n",
              "2           0  \n",
              "3           1  \n",
              "4           0  "
            ]
          },
          "execution_count": 275,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Aplicamos label encoder a la columna Profit_cat\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "le = LabelEncoder()\n",
        "subset['Profit_cat'] = le.fit_transform(subset['Profit_cat'])\n",
        "subset.head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 276,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 9994 entries, 0 to 9993\n",
            "Data columns (total 8 columns):\n",
            " #   Column       Non-Null Count  Dtype\n",
            "---  ------       --------------  -----\n",
            " 0   Express      9994 non-null   int64\n",
            " 1   First        9994 non-null   int64\n",
            " 2   Second       9994 non-null   int64\n",
            " 3   Standard     9994 non-null   int64\n",
            " 4   Consumidor   9994 non-null   int64\n",
            " 5   Corporativo  9994 non-null   int64\n",
            " 6   Oficina      9994 non-null   int64\n",
            " 7   Profit_cat   9994 non-null   int64\n",
            "dtypes: int64(8)\n",
            "memory usage: 624.8 KB\n"
          ]
        }
      ],
      "source": [
        "# Se transforma las columnas a tipo int\n",
        "subset = subset.astype(int)\n",
        "subset.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Aplicamos nuestras transformaciones a la data en una regresión logística"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 277,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Aplicamos una regresión logistica para predecir el valor de la columna Profit_cat\n",
        "import pandas as pd\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.linear_model import LogisticRegression, LinearRegression\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "from sklearn import metrics\n",
        "from sklearn.metrics import classification_report, roc_curve, roc_auc_score\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 278,
      "metadata": {},
      "outputs": [],
      "source": [
        "X = subset.drop('Profit_cat', axis=1).values  # Features (todas las columnas excepto la variable objetivo)\n",
        "y = subset['Profit_cat'].values  # Variable objetivo\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 279,
      "metadata": {},
      "outputs": [
        {
          "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>0</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Express</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>First</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Second</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Standard</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Consumidor</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Corporativo</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Oficina</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             0\n",
              "0      Express\n",
              "1        First\n",
              "2       Second\n",
              "3     Standard\n",
              "4   Consumidor\n",
              "5  Corporativo\n",
              "6      Oficina"
            ]
          },
          "execution_count": 279,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Vamos a conservar los nombres de las columnas para luego poder identificar los coeficientes\n",
        "X_columns = pd.DataFrame(subset.drop('Profit_cat', axis=1).columns)\n",
        "X_columns"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Estandarizamos la data\n",
        "* Usamos sklearn.preprocessing.StandardScaler ()\n",
        "* https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html\n",
        "* La estandarización es un proceso que se realiza sobre la data numérica.\n",
        "* La estandarización es un proceso mediante el cual se transforma la data para que tenga media 0 y desviación estándar 1.\n",
        "\n",
        "La fórmula es la siguiente:\n",
        "$z = (x - u)\\over s$\n",
        "\n",
        "Donde:\n",
        "* $z$: valor estandarizado\n",
        "* $x$: valor original\n",
        "* $u$: media\n",
        "* $s$: desviación estándar\n",
        "\n",
        "\n",
        "* Si necesitan escribir ecuaciones en markdown, pueden usar este sitio: https://www.codecogs.com/latex/eqneditor.php"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 280,
      "metadata": {},
      "outputs": [],
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "standard_x = StandardScaler()\n",
        "X_train = standard_x.fit_transform(X_train)\n",
        "X_test = standard_x.fit_transform(X_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 281,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Se crea el objeto\n",
        "logmodel = LogisticRegression(random_state=0) #Podemos usar distintos solver='lbfgs' como ejemplo\n",
        "\n",
        "# Se entrena el modelo\n",
        "modelo = logmodel.fit(X_train,y_train)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 282,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array([0, 0, 0, ..., 0, 0, 0])"
            ]
          },
          "execution_count": 282,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se realizan las predicciones\n",
        "predictions = logmodel.predict(X_test)\n",
        "predictions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 283,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(2999,)"
            ]
          },
          "execution_count": 283,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Se extraen las probabilidades del conjuto de test\n",
        "y_proba = logmodel.predict_proba(X_test)[:,1] # Se extrae la probabilidad en un array de 1 columna\n",
        "y_proba.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 284,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Podemos revisar manualmente las predicciones\n",
        "# Comparar las predicciones con los valores reales\n",
        "# predictions[0:10], y_test[0:10]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Extrayendo información del modelo"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 285,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Veamos el peso de cada variable en el modelo\n",
        "weights = pd.Series(modelo.coef_[0], index=X_columns[0])  # Hay que guardarlo como serie de pandas para poder graficarlo\n",
        "\n",
        "fig,axs = plt.subplots(1,2,figsize=(15,5))\n",
        "#plt.subplots_adjust(wspace=0.5)\n",
        "weights.sort_values(ascending=False).head(4).plot(kind='bar',ax=axs[0],\n",
        "                                                  title='Los 4 factores principales que aumentan en la ganancia',color='steelblue',\n",
        "                                                  fontsize=12)\n",
        "weights.sort_values(ascending=False).tail(4).plot(kind='bar',ax=axs[1],\n",
        "                                                  title='Los 4 factores principales que disminuyen la ganancia',color='tomato',\n",
        "                                                  fontsize=12);"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.10.4"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}