Aplicação_Regressão_Logistica.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOJF4fHFJs50TesyY/5XQsd",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/lauramsromano/logistic_regression_vendas/blob/main/Aplica%C3%A7%C3%A3o_Regress%C3%A3o_Logistica.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Regressão Logística:"
      ],
      "metadata": {
        "id": "-qM6UCWJlcvq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "> Feito por: Laura Maria de Souza Romano\n",
        "\n",
        "📧 e-mail - laura.sromano@hotmail.com\n",
        "🔗\n",
        "[linkedin](https://www.linkedin.com/in/laura-maria-de-souza-romano/)\n",
        "\n",
        "---"
      ],
      "metadata": {
        "id": "VPlC-zu7ixTT"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Com base no histórico de vendas que contém informações sobre vendas realizadas e canceladas, o objetivo é desenvolver um modelo de regressão logística para prever possíveis cancelamentos.\n",
        "\n",
        "A regressão logística é uma técnica de modelagem estatística usada para prever uma variável categórica binária, como a ocorrência ou não de cancelamentos. Nesse caso, a variável dependente será binária, indicando se uma venda foi cancelada (1) ou não (0), e as variáveis independentes podem incluir dados sobre o cliente, produto, contexto da venda, entre outros fatores relevantes.\n",
        "\n",
        "O modelo de regressão logística calculará as probabilidades de uma venda ser cancelada com base nas variáveis independentes e atribuirá uma classificação binária (0 ou 1) com base nessa probabilidade. Ao utilizar os dados históricos para treinar o modelo, ele aprenderá a encontrar padrões e relacionamentos entre as variáveis que possam influenciar a ocorrência de cancelamentos.\n",
        "\n",
        "Assim, após o treinamento do modelo, será possível fazer previsões sobre a probabilidade de uma venda ser cancelada com base nas informações disponíveis, permitindo que a empresa tome medidas preventivas ou proativas para evitar ou lidar adequadamente com possíveis cancelamentos.\n",
        "\n",
        "É fundamental garantir a qualidade dos dados de treinamento e a seleção adequada das variáveis independentes para obter um modelo de regressão logística eficaz. Além disso, a avaliação do desempenho do modelo usando métricas como a precisão, a sensibilidade (recall) e a especificidade é essencial para verificar sua acurácia e realizar ajustes, se necessário, para aprimorar suas previsões.\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "2-r_POQQmQI1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "> Base - https://archive.ics.uci.edu/dataset/352/online+retail"
      ],
      "metadata": {
        "id": "EnJSVymUZDfP"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Bibliotecas:"
      ],
      "metadata": {
        "id": "Ev_fvoLwws5E"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# bibliotecas padrões:\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "# import random\n",
        "\n",
        "# modelos de categorização e métricas:\n",
        "from sklearn.linear_model import LogisticRegression # regressão logistica\n",
        "from sklearn.neighbors import KNeighborsClassifier # KNN\n",
        "from sklearn.svm import SVC #SVM\n",
        "from sklearn.model_selection import train_test_split # separação treino e teste\n",
        "\n",
        "from sklearn.metrics import roc_auc_score #\n",
        "from sklearn.metrics import accuracy_score # cálculo da acurácia\n",
        "\n",
        "# pipeline:\n",
        "from sklearn.pipeline import Pipeline # gerando pipeline\n",
        "from sklearn.pipeline import make_pipeline # extraindo infos da pipeline\n",
        "from sklearn.compose import ColumnTransformer # agrupação de métodos para pipeline\n",
        "\n",
        "# processamento de variáveis numéricas:\n",
        "from sklearn.preprocessing import RobustScaler # ajuste da escala das features\n",
        "\n",
        "# processamento de variáveis nominais:\n",
        "!pip install category_encoders\n",
        "from category_encoders import CountEncoder # transformação por frequência\n",
        "from sklearn.impute import SimpleImputer #\n",
        "\n",
        "# processamento target:\n",
        "import imblearn\n",
        "from imblearn.under_sampling import RandomUnderSampler\n",
        "\n",
        "# validação cruzada:\n",
        "from sklearn.model_selection import (cross_validate, StratifiedKFold) # validação cruzada\n",
        "\n",
        "# tratamento data e hora\n",
        "from datetime import datetime\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "metadata": {
        "id": "-1dUWd1ZlD2Z",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8086fc04-95b2-48a3-b0fc-b770b6cc129e"
      },
      "execution_count": 86,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: category_encoders in /usr/local/lib/python3.10/dist-packages (2.6.1)\n",
            "Requirement already satisfied: numpy>=1.14.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.22.4)\n",
            "Requirement already satisfied: scikit-learn>=0.20.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.2.2)\n",
            "Requirement already satisfied: scipy>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.10.1)\n",
            "Requirement already satisfied: statsmodels>=0.9.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (0.13.5)\n",
            "Requirement already satisfied: pandas>=1.0.5 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.5.3)\n",
            "Requirement already satisfied: patsy>=0.5.1 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (0.5.3)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encoders) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encoders) (2022.7.1)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.10/dist-packages (from patsy>=0.5.1->category_encoders) (1.16.0)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.20.0->category_encoders) (1.3.1)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.20.0->category_encoders) (3.2.0)\n",
            "Requirement already satisfied: packaging>=21.3 in /usr/local/lib/python3.10/dist-packages (from statsmodels>=0.9.0->category_encoders) (23.1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# leitura da base original\n",
        "amostra = pd.read_excel('online_retail.xlsx')"
      ],
      "metadata": {
        "id": "V4AhZbxVlRNo"
      },
      "execution_count": 87,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### ***Análises iniciais da amostra original:***\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "dMv8Xvfr5OLX"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Resumo das análises:\n",
        "\n",
        "- Levantamento geral dos dados para tomada de decisão nas etapas de tratamento.\n",
        "  - Verificando divergências nas informações;  \n",
        "  - Levantamento percentual das informações verificadas;\n",
        "  - Verificando valores nulos;\n",
        "  - Outras informações presentes na amostra (períodos de tempo, paises presentes na base, etc.)"
      ],
      "metadata": {
        "id": "-ezakBac6H1Y"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "***Glossário de features já existentes na base:***\n",
        "- `InvoiceNo`: Número da fatura: número integral de 6 dígitos atribuído exclusivamente a cada transação.\n",
        "Se este código começar com a letra 'c', indica um cancelamento.\n",
        "\n",
        "  - Exemplo: C581490 - Cancelado, 581256 - Não Cancelado, A581490, 256, etc.\t- Código de transação inválido\n",
        "\n",
        "- `StockCode`: Código do produto: um número integral de 5 dígitos atribuído exclusivamente a cada produto distinto.\n",
        "\n",
        "  - Exemplo: 35004 - Código de produto válido, 35004C, M, etc. - Código de produto inválido.\n",
        "\n",
        "- `Description`: Nome do produto.\n",
        "-` Quantity`: As quantidades de cada produto (item) por transação.\n",
        "- `InvoiceDate`: Dia e a hora em que cada transação foi gerada.\n",
        "- `UnitPrice`: Preço do produto por unidade em libras esterlinas.\n",
        "- `CustomerID`: Número integral de 5 dígitos atribuído exclusivamente a cada cliente.\n",
        "\n",
        "  - Exemplo: 14527 - Código de cliente válido, F14527, 855, etc. - Código de cliente inválido.\n",
        "\n",
        "- `Country`: Nome do país onde cada cliente reside.\n",
        "\n",
        "***Features adicionadas durante o tratamento:***\n",
        "\n",
        "- `TotalPrice`: Valor total de cada transação, resultado da multiplicação entre as colunas `'UnitPrice'` e `'Quantity'`.\n",
        "- `Status`: Informação se a transação foi cancelada (Cancelado) ou não (Não Cancelado)."
      ],
      "metadata": {
        "id": "ICqnBBf5u1xe"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Amostra de dados:\n",
        "amostra.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "LHsXYk_ms-MA",
        "outputId": "ced95453-c5d1-4016-bed7-2edc83ec0082"
      },
      "execution_count": 88,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  InvoiceNo StockCode                          Description  Quantity  \\\n",
              "0    536365    85123A   WHITE HANGING HEART T-LIGHT HOLDER         6   \n",
              "1    536365     71053                  WHITE METAL LANTERN         6   \n",
              "2    536365    84406B       CREAM CUPID HEARTS COAT HANGER         8   \n",
              "3    536365    84029G  KNITTED UNION FLAG HOT WATER BOTTLE         6   \n",
              "4    536365    84029E       RED WOOLLY HOTTIE WHITE HEART.         6   \n",
              "\n",
              "          InvoiceDate  UnitPrice  CustomerID         Country  \n",
              "0 2010-12-01 08:26:00       2.55     17850.0  United Kingdom  \n",
              "1 2010-12-01 08:26:00       3.39     17850.0  United Kingdom  \n",
              "2 2010-12-01 08:26:00       2.75     17850.0  United Kingdom  \n",
              "3 2010-12-01 08:26:00       3.39     17850.0  United Kingdom  \n",
              "4 2010-12-01 08:26:00       3.39     17850.0  United Kingdom  "
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-2f4ceb61-9e1a-4f84-8026-7705eadbb612\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>InvoiceNo</th>\n",
              "      <th>StockCode</th>\n",
              "      <th>Description</th>\n",
              "      <th>Quantity</th>\n",
              "      <th>InvoiceDate</th>\n",
              "      <th>UnitPrice</th>\n",
              "      <th>CustomerID</th>\n",
              "      <th>Country</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>536365</td>\n",
              "      <td>85123A</td>\n",
              "      <td>WHITE HANGING HEART T-LIGHT HOLDER</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>2.55</td>\n",
              "      <td>17850.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>536365</td>\n",
              "      <td>71053</td>\n",
              "      <td>WHITE METAL LANTERN</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>3.39</td>\n",
              "      <td>17850.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>536365</td>\n",
              "      <td>84406B</td>\n",
              "      <td>CREAM CUPID HEARTS COAT HANGER</td>\n",
              "      <td>8</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>2.75</td>\n",
              "      <td>17850.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>536365</td>\n",
              "      <td>84029G</td>\n",
              "      <td>KNITTED UNION FLAG HOT WATER BOTTLE</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>3.39</td>\n",
              "      <td>17850.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>536365</td>\n",
              "      <td>84029E</td>\n",
              "      <td>RED WOOLLY HOTTIE WHITE HEART.</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>3.39</td>\n",
              "      <td>17850.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2f4ceb61-9e1a-4f84-8026-7705eadbb612')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-86a95968-b62a-492c-8ce2-0f8bbf92b9ea\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-86a95968-b62a-492c-8ce2-0f8bbf92b9ea')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-86a95968-b62a-492c-8ce2-0f8bbf92b9ea button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-2f4ceb61-9e1a-4f84-8026-7705eadbb612 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-2f4ceb61-9e1a-4f84-8026-7705eadbb612');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 88
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# infos gerais:\n",
        "amostra.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6G96U2_gINna",
        "outputId": "2df0df49-7f8e-4b57-c826-2d078a78d656"
      },
      "execution_count": 89,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 541909 entries, 0 to 541908\n",
            "Data columns (total 8 columns):\n",
            " #   Column       Non-Null Count   Dtype         \n",
            "---  ------       --------------   -----         \n",
            " 0   InvoiceNo    541909 non-null  object        \n",
            " 1   StockCode    541909 non-null  object        \n",
            " 2   Description  540455 non-null  object        \n",
            " 3   Quantity     541909 non-null  int64         \n",
            " 4   InvoiceDate  541909 non-null  datetime64[ns]\n",
            " 5   UnitPrice    541909 non-null  float64       \n",
            " 6   CustomerID   406829 non-null  float64       \n",
            " 7   Country      541909 non-null  object        \n",
            "dtypes: datetime64[ns](1), float64(2), int64(1), object(4)\n",
            "memory usage: 33.1+ MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "amostra.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oJ5PwXNpsxsT",
        "outputId": "3ebabcab-2c84-4986-a051-e20aab351887"
      },
      "execution_count": 90,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(541909, 8)"
            ]
          },
          "metadata": {},
          "execution_count": 90
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# informações estatísticas:\n",
        "amostra.describe().T"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "tMrYC8TSyKQu",
        "outputId": "1030a08e-5c01-4278-fbd6-8b3f6e6e8e84"
      },
      "execution_count": 91,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               count          mean          std       min       25%       50%  \\\n",
              "Quantity    541909.0      9.552250   218.081158 -80995.00      1.00      3.00   \n",
              "UnitPrice   541909.0      4.611114    96.759853 -11062.06      1.25      2.08   \n",
              "CustomerID  406829.0  15287.690570  1713.600303  12346.00  13953.00  15152.00   \n",
              "\n",
              "                 75%      max  \n",
              "Quantity       10.00  80995.0  \n",
              "UnitPrice       4.13  38970.0  \n",
              "CustomerID  16791.00  18287.0  "
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-560d8c52-3c72-4d3d-863b-257f1aebf5a9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>count</th>\n",
              "      <th>mean</th>\n",
              "      <th>std</th>\n",
              "      <th>min</th>\n",
              "      <th>25%</th>\n",
              "      <th>50%</th>\n",
              "      <th>75%</th>\n",
              "      <th>max</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Quantity</th>\n",
              "      <td>541909.0</td>\n",
              "      <td>9.552250</td>\n",
              "      <td>218.081158</td>\n",
              "      <td>-80995.00</td>\n",
              "      <td>1.00</td>\n",
              "      <td>3.00</td>\n",
              "      <td>10.00</td>\n",
              "      <td>80995.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>UnitPrice</th>\n",
              "      <td>541909.0</td>\n",
              "      <td>4.611114</td>\n",
              "      <td>96.759853</td>\n",
              "      <td>-11062.06</td>\n",
              "      <td>1.25</td>\n",
              "      <td>2.08</td>\n",
              "      <td>4.13</td>\n",
              "      <td>38970.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>CustomerID</th>\n",
              "      <td>406829.0</td>\n",
              "      <td>15287.690570</td>\n",
              "      <td>1713.600303</td>\n",
              "      <td>12346.00</td>\n",
              "      <td>13953.00</td>\n",
              "      <td>15152.00</td>\n",
              "      <td>16791.00</td>\n",
              "      <td>18287.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-560d8c52-3c72-4d3d-863b-257f1aebf5a9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-a89af08f-c3cd-46ac-84f7-b7bd0027dd0b\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-a89af08f-c3cd-46ac-84f7-b7bd0027dd0b')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-a89af08f-c3cd-46ac-84f7-b7bd0027dd0b button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-560d8c52-3c72-4d3d-863b-257f1aebf5a9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-560d8c52-3c72-4d3d-863b-257f1aebf5a9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 91
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Foi observado que as colunas `'Quantity' `e `'UnitPrice'` apresentaram **valores negativos**, porém, esses valores correspondem a compras canceladas e não serão tratados ou removidos da base de dados.\n",
        "\n",
        "- Além disso, identificamos que tanto a coluna `'Quantity'` quanto a coluna `'UnitPrice'` apresentam um **desvio padrão elevado**, o que pode impactar negativamente o desempenho do modelo. Para lidar com essa situação iremos explorar alternativas que visam melhorar a distribuição dos dados."
      ],
      "metadata": {
        "id": "y-vsC9XdyZPh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# vendas canceladas:\n",
        "amostra['InvoiceNo'] = amostra['InvoiceNo'].astype(str)\n",
        "amostra[amostra['InvoiceNo'].str.contains('C')]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "w126PHVUyM_-",
        "outputId": "d49d298d-37b8-4be1-9a75-9c10d5d22921"
      },
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       InvoiceNo StockCode                       Description  Quantity  \\\n",
              "141      C536379         D                          Discount        -1   \n",
              "154      C536383    35004C   SET OF 3 COLOURED  FLYING DUCKS        -1   \n",
              "235      C536391     22556    PLASTERS IN TIN CIRCUS PARADE        -12   \n",
              "236      C536391     21984  PACK OF 12 PINK PAISLEY TISSUES        -24   \n",
              "237      C536391     21983  PACK OF 12 BLUE PAISLEY TISSUES        -24   \n",
              "...          ...       ...                               ...       ...   \n",
              "540449   C581490     23144   ZINC T-LIGHT HOLDER STARS SMALL       -11   \n",
              "541541   C581499         M                            Manual        -1   \n",
              "541715   C581568     21258        VICTORIAN SEWING BOX LARGE        -5   \n",
              "541716   C581569     84978  HANGING HEART JAR T-LIGHT HOLDER        -1   \n",
              "541717   C581569     20979     36 PENCILS TUBE RED RETROSPOT        -5   \n",
              "\n",
              "               InvoiceDate  UnitPrice  CustomerID         Country  \n",
              "141    2010-12-01 09:41:00      27.50     14527.0  United Kingdom  \n",
              "154    2010-12-01 09:49:00       4.65     15311.0  United Kingdom  \n",
              "235    2010-12-01 10:24:00       1.65     17548.0  United Kingdom  \n",
              "236    2010-12-01 10:24:00       0.29     17548.0  United Kingdom  \n",
              "237    2010-12-01 10:24:00       0.29     17548.0  United Kingdom  \n",
              "...                    ...        ...         ...             ...  \n",
              "540449 2011-12-09 09:57:00       0.83     14397.0  United Kingdom  \n",
              "541541 2011-12-09 10:28:00     224.69     15498.0  United Kingdom  \n",
              "541715 2011-12-09 11:57:00      10.95     15311.0  United Kingdom  \n",
              "541716 2011-12-09 11:58:00       1.25     17315.0  United Kingdom  \n",
              "541717 2011-12-09 11:58:00       1.25     17315.0  United Kingdom  \n",
              "\n",
              "[9288 rows x 8 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-73196975-5151-49c8-8538-93f4dba67ecc\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>InvoiceNo</th>\n",
              "      <th>StockCode</th>\n",
              "      <th>Description</th>\n",
              "      <th>Quantity</th>\n",
              "      <th>InvoiceDate</th>\n",
              "      <th>UnitPrice</th>\n",
              "      <th>CustomerID</th>\n",
              "      <th>Country</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>141</th>\n",
              "      <td>C536379</td>\n",
              "      <td>D</td>\n",
              "      <td>Discount</td>\n",
              "      <td>-1</td>\n",
              "      <td>2010-12-01 09:41:00</td>\n",
              "      <td>27.50</td>\n",
              "      <td>14527.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>154</th>\n",
              "      <td>C536383</td>\n",
              "      <td>35004C</td>\n",
              "      <td>SET OF 3 COLOURED  FLYING DUCKS</td>\n",
              "      <td>-1</td>\n",
              "      <td>2010-12-01 09:49:00</td>\n",
              "      <td>4.65</td>\n",
              "      <td>15311.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>235</th>\n",
              "      <td>C536391</td>\n",
              "      <td>22556</td>\n",
              "      <td>PLASTERS IN TIN CIRCUS PARADE</td>\n",
              "      <td>-12</td>\n",
              "      <td>2010-12-01 10:24:00</td>\n",
              "      <td>1.65</td>\n",
              "      <td>17548.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>236</th>\n",
              "      <td>C536391</td>\n",
              "      <td>21984</td>\n",
              "      <td>PACK OF 12 PINK PAISLEY TISSUES</td>\n",
              "      <td>-24</td>\n",
              "      <td>2010-12-01 10:24:00</td>\n",
              "      <td>0.29</td>\n",
              "      <td>17548.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>237</th>\n",
              "      <td>C536391</td>\n",
              "      <td>21983</td>\n",
              "      <td>PACK OF 12 BLUE PAISLEY TISSUES</td>\n",
              "      <td>-24</td>\n",
              "      <td>2010-12-01 10:24:00</td>\n",
              "      <td>0.29</td>\n",
              "      <td>17548.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>540449</th>\n",
              "      <td>C581490</td>\n",
              "      <td>23144</td>\n",
              "      <td>ZINC T-LIGHT HOLDER STARS SMALL</td>\n",
              "      <td>-11</td>\n",
              "      <td>2011-12-09 09:57:00</td>\n",
              "      <td>0.83</td>\n",
              "      <td>14397.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541541</th>\n",
              "      <td>C581499</td>\n",
              "      <td>M</td>\n",
              "      <td>Manual</td>\n",
              "      <td>-1</td>\n",
              "      <td>2011-12-09 10:28:00</td>\n",
              "      <td>224.69</td>\n",
              "      <td>15498.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541715</th>\n",
              "      <td>C581568</td>\n",
              "      <td>21258</td>\n",
              "      <td>VICTORIAN SEWING BOX LARGE</td>\n",
              "      <td>-5</td>\n",
              "      <td>2011-12-09 11:57:00</td>\n",
              "      <td>10.95</td>\n",
              "      <td>15311.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541716</th>\n",
              "      <td>C581569</td>\n",
              "      <td>84978</td>\n",
              "      <td>HANGING HEART JAR T-LIGHT HOLDER</td>\n",
              "      <td>-1</td>\n",
              "      <td>2011-12-09 11:58:00</td>\n",
              "      <td>1.25</td>\n",
              "      <td>17315.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541717</th>\n",
              "      <td>C581569</td>\n",
              "      <td>20979</td>\n",
              "      <td>36 PENCILS TUBE RED RETROSPOT</td>\n",
              "      <td>-5</td>\n",
              "      <td>2011-12-09 11:58:00</td>\n",
              "      <td>1.25</td>\n",
              "      <td>17315.0</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>9288 rows × 8 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-73196975-5151-49c8-8538-93f4dba67ecc')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-fe042e9a-2038-4f67-84e9-6f57288ad863\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-fe042e9a-2038-4f67-84e9-6f57288ad863')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-fe042e9a-2038-4f67-84e9-6f57288ad863 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-73196975-5151-49c8-8538-93f4dba67ecc button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-73196975-5151-49c8-8538-93f4dba67ecc');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 92
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# qtd. de vendas canceladas:\n",
        "len(amostra[amostra['InvoiceNo'].str.contains('C')])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1g2DT1cj9nBF",
        "outputId": "9138330b-4abc-4166-dcf3-3b8b6784cbd6"
      },
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "9288"
            ]
          },
          "metadata": {},
          "execution_count": 93
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# verificando se há valores negativos em 'Quantity' ou em 'UnitPrice' que não sejam referentes a vendas canceladas:\n",
        "amostra[(~amostra['InvoiceNo'].str.startswith('C')) & ((amostra['Quantity'] <= 0) | (amostra['UnitPrice'] <= 0))]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "IDx4kqM3EXG1",
        "outputId": "baee92e4-2546-4a2e-d924-fe724c6d9a18"
      },
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       InvoiceNo StockCode                   Description  Quantity  \\\n",
              "622       536414     22139                           NaN        56   \n",
              "1970      536545     21134                           NaN         1   \n",
              "1971      536546     22145                           NaN         1   \n",
              "1972      536547     37509                           NaN         1   \n",
              "1987      536549    85226A                           NaN         1   \n",
              "...          ...       ...                           ...       ...   \n",
              "536981    581234     72817                           NaN        27   \n",
              "538504    581406    46000M  POLYESTER FILLER PAD 45x45cm       240   \n",
              "538505    581406    46000S  POLYESTER FILLER PAD 40x40cm       300   \n",
              "538554    581408     85175                           NaN        20   \n",
              "538919    581422     23169                       smashed      -235   \n",
              "\n",
              "               InvoiceDate  UnitPrice  CustomerID         Country  \n",
              "622    2010-12-01 11:52:00        0.0         NaN  United Kingdom  \n",
              "1970   2010-12-01 14:32:00        0.0         NaN  United Kingdom  \n",
              "1971   2010-12-01 14:33:00        0.0         NaN  United Kingdom  \n",
              "1972   2010-12-01 14:33:00        0.0         NaN  United Kingdom  \n",
              "1987   2010-12-01 14:34:00        0.0         NaN  United Kingdom  \n",
              "...                    ...        ...         ...             ...  \n",
              "536981 2011-12-08 10:33:00        0.0         NaN  United Kingdom  \n",
              "538504 2011-12-08 13:58:00        0.0         NaN  United Kingdom  \n",
              "538505 2011-12-08 13:58:00        0.0         NaN  United Kingdom  \n",
              "538554 2011-12-08 14:06:00        0.0         NaN  United Kingdom  \n",
              "538919 2011-12-08 15:24:00        0.0         NaN  United Kingdom  \n",
              "\n",
              "[2517 rows x 8 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-a1696569-8f15-47d3-9c39-be7fb632606f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>InvoiceNo</th>\n",
              "      <th>StockCode</th>\n",
              "      <th>Description</th>\n",
              "      <th>Quantity</th>\n",
              "      <th>InvoiceDate</th>\n",
              "      <th>UnitPrice</th>\n",
              "      <th>CustomerID</th>\n",
              "      <th>Country</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>622</th>\n",
              "      <td>536414</td>\n",
              "      <td>22139</td>\n",
              "      <td>NaN</td>\n",
              "      <td>56</td>\n",
              "      <td>2010-12-01 11:52:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1970</th>\n",
              "      <td>536545</td>\n",
              "      <td>21134</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>2010-12-01 14:32:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1971</th>\n",
              "      <td>536546</td>\n",
              "      <td>22145</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>2010-12-01 14:33:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1972</th>\n",
              "      <td>536547</td>\n",
              "      <td>37509</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>2010-12-01 14:33:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1987</th>\n",
              "      <td>536549</td>\n",
              "      <td>85226A</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>2010-12-01 14:34:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>536981</th>\n",
              "      <td>581234</td>\n",
              "      <td>72817</td>\n",
              "      <td>NaN</td>\n",
              "      <td>27</td>\n",
              "      <td>2011-12-08 10:33:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>538504</th>\n",
              "      <td>581406</td>\n",
              "      <td>46000M</td>\n",
              "      <td>POLYESTER FILLER PAD 45x45cm</td>\n",
              "      <td>240</td>\n",
              "      <td>2011-12-08 13:58:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>538505</th>\n",
              "      <td>581406</td>\n",
              "      <td>46000S</td>\n",
              "      <td>POLYESTER FILLER PAD 40x40cm</td>\n",
              "      <td>300</td>\n",
              "      <td>2011-12-08 13:58:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>538554</th>\n",
              "      <td>581408</td>\n",
              "      <td>85175</td>\n",
              "      <td>NaN</td>\n",
              "      <td>20</td>\n",
              "      <td>2011-12-08 14:06:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>538919</th>\n",
              "      <td>581422</td>\n",
              "      <td>23169</td>\n",
              "      <td>smashed</td>\n",
              "      <td>-235</td>\n",
              "      <td>2011-12-08 15:24:00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2517 rows × 8 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a1696569-8f15-47d3-9c39-be7fb632606f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-e2d0c23d-5d96-47a9-98e4-58f688f209ae\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-e2d0c23d-5d96-47a9-98e4-58f688f209ae')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-e2d0c23d-5d96-47a9-98e4-58f688f209ae button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a1696569-8f15-47d3-9c39-be7fb632606f button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a1696569-8f15-47d3-9c39-be7fb632606f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 94
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# percentual de vendas 'não canceladas' com qtd ou valor unitário negativos:\n",
        "round(len(amostra[(~amostra['InvoiceNo'].str.startswith('C')) & ((amostra['Quantity'] <= 0) | (amostra['UnitPrice'] <= 0))])/ len(amostra)*100, 2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kp7ac8NxY98F",
        "outputId": "8cfdebcc-4259-4b96-f7cb-cd56d69a45f2"
      },
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.46"
            ]
          },
          "metadata": {},
          "execution_count": 95
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# percentual de vendas canceladas:\n",
        "round(len(amostra[amostra['InvoiceNo'].str.contains('C')]) / len(amostra)*100, 2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rKWd4eZ5DdsG",
        "outputId": "ddc4bf87-02c4-4da5-cbda-9afbe83e09d1"
      },
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.71"
            ]
          },
          "metadata": {},
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Aproximadamente **0.5%** dos valores na base apresentam informações zeradas ou negativas em `'Quantity'` e `'UnitPrice'`, porém, não estão relacionados a vendas canceladas. Para garantir a consistência dos dados, iremos proceder com a remoção dessas informações na base.\n",
        "\n",
        "- O percentual de vendas canceladas na amostra é de aproximadamente **2%** dos dados, indicando um desequilíbrio nos dados. Para abordar essa questão, iremos verificar e aplicar técnicas para realizar o balancear os dados."
      ],
      "metadata": {
        "id": "UW1ZOgo-DhfO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# verificando valores ausentes:\n",
        "(amostra.isna().sum() / len(amostra)*100).sort_values(ascending=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cPpkn6TVITWA",
        "outputId": "bb2ed200-6dae-43b8-aafa-df43edcaaa0e"
      },
      "execution_count": 97,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CustomerID     24.926694\n",
              "Description     0.268311\n",
              "InvoiceNo       0.000000\n",
              "StockCode       0.000000\n",
              "Quantity        0.000000\n",
              "InvoiceDate     0.000000\n",
              "UnitPrice       0.000000\n",
              "Country         0.000000\n",
              "dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 97
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# verificando se há valores diferentes do padrão para 'InvoiceNo' (valores devem ter 6 ou 7 digitos [para casos cancelado])\n",
        "codigos_unicos_invoice = amostra['InvoiceNo'].astype(str).tolist()\n",
        "codigos_invalidos = [codigo for codigo in codigos_unicos_invoice if len(str(codigo)) < 6 or len(str(codigo)) > 7]\n",
        "f'Há {len(codigos_invalidos)} código(s) divergente(s). Alguns exemplos: {codigos_invalidos}'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "GhhxvLfLgk5W",
        "outputId": "89d8a17b-8fb7-4c3d-e30e-39e81de0b374"
      },
      "execution_count": 98,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Há 0 código(s) divergente(s). Alguns exemplos: []'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 98
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# verificando se há valores diferentes do padrão para 'InvoiceNo' (valores devem ter 5 dígitos)\n",
        "codigos_unicos_stock = amostra['StockCode'].tolist()\n",
        "codigos_invalidos = [codigo for codigo in codigos_unicos_stock if len(str(codigo)) != 5 or not str(codigo).isdigit()]\n",
        "f'Há {len(codigos_invalidos)} código(s) divergente(s). Alguns exemplos: {codigos_invalidos[:5]}'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "pbSzBPCUiZOu",
        "outputId": "19a74f05-5e6e-4b9e-b0e5-2bfc05887f70"
      },
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "\"Há 54873 código(s) divergente(s). Alguns exemplos: ['85123A', '84406B', '84029G', '84029E', 'POST']\""
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 99
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "round(len(codigos_invalidos) / len(amostra)*100,2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NynplxKnqZ1l",
        "outputId": "881f80ea-1fd1-460e-8530-9b358b5b916c"
      },
      "execution_count": 100,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10.13"
            ]
          },
          "metadata": {},
          "execution_count": 100
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# alterando o tipo da variável 'CustomerID' para string e removendo o '.0' no final:\n",
        "amostra['CustomerID'] = amostra['CustomerID'].astype(str).str.split('.').str[0]"
      ],
      "metadata": {
        "id": "5V2d8JO5wU1S"
      },
      "execution_count": 101,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# verificando se há valores diferentes do padrão para 'CustomerID'\n",
        "codigos_unicos_customer = amostra['CustomerID'].tolist()\n",
        "codigos_invalidos = [codigo for codigo in codigos_unicos_customer if len(str(codigo)) != 5 or not str(codigo).isdigit()]\n",
        "f'Há {len(codigos_invalidos)} código(s) divergente(s). Alguns exemplos: {codigos_invalidos[:5]}'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "ck-J21QvkX1V",
        "outputId": "d1e45761-1350-48ce-ec8e-f8d96ff8a25d"
      },
      "execution_count": 102,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "\"Há 135080 código(s) divergente(s). Alguns exemplos: ['nan', 'nan', 'nan', 'nan', 'nan']\""
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 102
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "round(len(codigos_invalidos) / len(amostra)*100,2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ufdrf7vODwQl",
        "outputId": "b8dd1df9-33f2-4066-9d3f-1c1adebd058a"
      },
      "execution_count": 103,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "24.93"
            ]
          },
          "metadata": {},
          "execution_count": 103
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- A coluna `'CustomerID'` apresenta valores nulos, representando aproximadamente **25%** dos registros. Essa coluna é importante para o poder de predição do modelo, portanto será necessário encontrar uma estratégia adequada para lidar com esses valores ausentes.\n",
        "\n",
        "- A coluna `'InvoiceNo'` não apresenta valores despadronizados, ou seja, não há problemas a serem tratados nessa coluna.\n",
        "\n",
        "- A coluna `'StockCode'` possui valores despadronizados, afetando cerca de **10%** dos registros. Será necessário realizar algum tipo de tratamento para lidar com esses valores inconsistentes.\n",
        "\n",
        "- Os valores despadronizados em `'CustomerID'` são registros em 'nan', representando **24%** dos dados. Como pontuado, essa coluna é importante para o poder de predição do modelo, portanto será necessário encontrar uma estratégia adequada para lidar com esses valores divergentes."
      ],
      "metadata": {
        "id": "3wqKRWeu5WUa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Verificando proporção de infos por pais:\n",
        "amostra['Country'].value_counts() / len(amostra)*100"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q6FDZat-Uf-S",
        "outputId": "d0b17e9f-6301-47b6-801d-a8466c1c11fe"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "United Kingdom          91.431956\n",
              "Germany                  1.752139\n",
              "France                   1.579047\n",
              "EIRE                     1.512431\n",
              "Spain                    0.467422\n",
              "Netherlands              0.437527\n",
              "Belgium                  0.381798\n",
              "Switzerland              0.369435\n",
              "Portugal                 0.280305\n",
              "Australia                0.232327\n",
              "Norway                   0.200403\n",
              "Italy                    0.148180\n",
              "Channel Islands          0.139876\n",
              "Finland                  0.128250\n",
              "Cyprus                   0.114779\n",
              "Sweden                   0.085254\n",
              "Unspecified              0.082302\n",
              "Austria                  0.073998\n",
              "Denmark                  0.071783\n",
              "Japan                    0.066063\n",
              "Poland                   0.062926\n",
              "Israel                   0.054806\n",
              "USA                      0.053699\n",
              "Hong Kong                0.053145\n",
              "Singapore                0.042258\n",
              "Iceland                  0.033585\n",
              "Canada                   0.027864\n",
              "Greece                   0.026942\n",
              "Malta                    0.023436\n",
              "United Arab Emirates     0.012548\n",
              "European Community       0.011257\n",
              "RSA                      0.010703\n",
              "Lebanon                  0.008304\n",
              "Lithuania                0.006459\n",
              "Brazil                   0.005905\n",
              "Czech Republic           0.005536\n",
              "Bahrain                  0.003506\n",
              "Saudi Arabia             0.001845\n",
              "Name: Country, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 104
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Qtd de paises:\n",
        "amostra['Country'].nunique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VlrAu_oo22z7",
        "outputId": "008d6c8b-2e3f-45c9-95e9-1a2051c82387"
      },
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "38"
            ]
          },
          "metadata": {},
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- 91% dos dados são referentes a compras realizadas no **Reino Unido (United Kingdom**), o que indica uma concentração significativa de registros nesse país. Ou seja, as informação pode ser relevante para **análises específicas relacionadas ao mercado britânico.**"
      ],
      "metadata": {
        "id": "SCELiwywdcpA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Anos presente na base:\n",
        "amostra['InvoiceDate'].dt.year.unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cgzlbI_PWFEu",
        "outputId": "54681c7e-c5f8-4869-b2d5-3a27f14a6974"
      },
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2010, 2011])"
            ]
          },
          "metadata": {},
          "execution_count": 106
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Qtd de itens:\n",
        "amostra['StockCode'].nunique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MtFWdgSfWVsh",
        "outputId": "d8dade93-9864-4969-c957-936399a08e2b"
      },
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4070"
            ]
          },
          "metadata": {},
          "execution_count": 107
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Qtd de clientes:\n",
        "amostra['CustomerID'].dropna().nunique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cYw1av5A4IDp",
        "outputId": "c461c2a0-85b8-4a1e-a1f7-b8d231ae5fc5"
      },
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4373"
            ]
          },
          "metadata": {},
          "execution_count": 108
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### **Resumo das conclusões iniciais:**\n",
        "\n",
        "A amostra contém dados de vendas referentes aos anos de 2010 e 2011. Esses dados incluem um total de 4.070 itens diferentes vendidos para um conjunto de 4.373 clientes em 38 países, sendo United Kingdom (Reino Unido) o pais mais representativo, com uma parcela de 91% da amostra.\n",
        "\n",
        "> Colunas `'Quantity'` e `'UnitPrice'`:\n",
        "\n",
        "Os valores negativos encontrados nessas colunas são referentes a vendas canceladas. Para evitar qualquer influência indesejada na base de dados, as vendas não canceladas com informações negativas ou zeradas nessas colunas serão removidas.\n",
        "\n",
        "Além disso, é importante destacar que as colunas `'Quantity'` e `'UnitPrice'` apresentam um desvio padrão elevado, o que pode prejudicar o desempenho do modelo. A fim de melhorar a distribuição dos dados, serão exploradas alternativas para tratar essa situação.\n",
        "\n",
        "> Coluna `'InvoiceNo'`:\n",
        "\n",
        "Observamos que a base de dados está desbalanceada, com apenas 2% dos registros relacionados a cancelamentos. Para ajustar esse desequilíbrio, será necessário aplicar técnicas apropriadas de tratamento, visando obter uma distribuição mais equitativa e representativa dos dados.\n",
        "\n",
        "> Colunas `'CustomerID'` e '`StockCode'`:\n",
        "\n",
        "Ambas as colunas apresentam valores despadronizados que precisarão ser tratados."
      ],
      "metadata": {
        "id": "ezB_hyA71w3b"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### ***Preparação e análise da base:***\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "AGujc1zj507l"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Resumo dos passos:\n",
        "\n",
        "- Gerando a base que será utilizada e fazendo os ajustes necessários:\n",
        "  - Tratamento de informações divergentes;\n",
        "  - Tratamento de nulos;\n",
        "  - Filtros.\n",
        "- Tratamento, criação e análise das features.\n",
        "- Definição, criação, tratamento e análise da target.\n",
        "- EDA das informações finais da base."
      ],
      "metadata": {
        "id": "ytK7sAOgTnf7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Criando base para tratamento e aplicação do modelo, matendo a 'amostra' sem alterações:\n",
        "base_amostra = pd.DataFrame()\n",
        "base_amostra = amostra.copy()\n",
        "base_amostra"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "PdR8ljOfGQSl",
        "outputId": "9eb20c85-2f25-46cd-a54e-1ad777804525"
      },
      "execution_count": 109,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       InvoiceNo StockCode                          Description  Quantity  \\\n",
              "0         536365    85123A   WHITE HANGING HEART T-LIGHT HOLDER         6   \n",
              "1         536365     71053                  WHITE METAL LANTERN         6   \n",
              "2         536365    84406B       CREAM CUPID HEARTS COAT HANGER         8   \n",
              "3         536365    84029G  KNITTED UNION FLAG HOT WATER BOTTLE         6   \n",
              "4         536365    84029E       RED WOOLLY HOTTIE WHITE HEART.         6   \n",
              "...          ...       ...                                  ...       ...   \n",
              "541904    581587     22613          PACK OF 20 SPACEBOY NAPKINS        12   \n",
              "541905    581587     22899         CHILDREN'S APRON DOLLY GIRL          6   \n",
              "541906    581587     23254        CHILDRENS CUTLERY DOLLY GIRL          4   \n",
              "541907    581587     23255      CHILDRENS CUTLERY CIRCUS PARADE         4   \n",
              "541908    581587     22138        BAKING SET 9 PIECE RETROSPOT          3   \n",
              "\n",
              "               InvoiceDate  UnitPrice CustomerID         Country  \n",
              "0      2010-12-01 08:26:00       2.55      17850  United Kingdom  \n",
              "1      2010-12-01 08:26:00       3.39      17850  United Kingdom  \n",
              "2      2010-12-01 08:26:00       2.75      17850  United Kingdom  \n",
              "3      2010-12-01 08:26:00       3.39      17850  United Kingdom  \n",
              "4      2010-12-01 08:26:00       3.39      17850  United Kingdom  \n",
              "...                    ...        ...        ...             ...  \n",
              "541904 2011-12-09 12:50:00       0.85      12680          France  \n",
              "541905 2011-12-09 12:50:00       2.10      12680          France  \n",
              "541906 2011-12-09 12:50:00       4.15      12680          France  \n",
              "541907 2011-12-09 12:50:00       4.15      12680          France  \n",
              "541908 2011-12-09 12:50:00       4.95      12680          France  \n",
              "\n",
              "[541909 rows x 8 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-310e9e81-7f95-4aa3-9c43-ba8d24c198b8\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>InvoiceNo</th>\n",
              "      <th>StockCode</th>\n",
              "      <th>Description</th>\n",
              "      <th>Quantity</th>\n",
              "      <th>InvoiceDate</th>\n",
              "      <th>UnitPrice</th>\n",
              "      <th>CustomerID</th>\n",
              "      <th>Country</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>536365</td>\n",
              "      <td>85123A</td>\n",
              "      <td>WHITE HANGING HEART T-LIGHT HOLDER</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>2.55</td>\n",
              "      <td>17850</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>536365</td>\n",
              "      <td>71053</td>\n",
              "      <td>WHITE METAL LANTERN</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>3.39</td>\n",
              "      <td>17850</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>536365</td>\n",
              "      <td>84406B</td>\n",
              "      <td>CREAM CUPID HEARTS COAT HANGER</td>\n",
              "      <td>8</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>2.75</td>\n",
              "      <td>17850</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>536365</td>\n",
              "      <td>84029G</td>\n",
              "      <td>KNITTED UNION FLAG HOT WATER BOTTLE</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>3.39</td>\n",
              "      <td>17850</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>536365</td>\n",
              "      <td>84029E</td>\n",
              "      <td>RED WOOLLY HOTTIE WHITE HEART.</td>\n",
              "      <td>6</td>\n",
              "      <td>2010-12-01 08:26:00</td>\n",
              "      <td>3.39</td>\n",
              "      <td>17850</td>\n",
              "      <td>United Kingdom</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541904</th>\n",
              "      <td>581587</td>\n",
              "      <td>22613</td>\n",
              "      <td>PACK OF 20 SPACEBOY NAPKINS</td>\n",
              "      <td>12</td>\n",
              "      <td>2011-12-09 12:50:00</td>\n",
              "      <td>0.85</td>\n",
              "      <td>12680</td>\n",
              "      <td>France</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541905</th>\n",
              "      <td>581587</td>\n",
              "      <td>22899</td>\n",
              "      <td>CHILDREN'S APRON DOLLY GIRL</td>\n",
              "      <td>6</td>\n",
              "      <td>2011-12-09 12:50:00</td>\n",
              "      <td>2.10</td>\n",
              "      <td>12680</td>\n",
              "      <td>France</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541906</th>\n",
              "      <td>581587</td>\n",
              "      <td>23254</td>\n",
              "      <td>CHILDRENS CUTLERY DOLLY GIRL</td>\n",
              "      <td>4</td>\n",
              "      <td>2011-12-09 12:50:00</td>\n",
              "      <td>4.15</td>\n",
              "      <td>12680</td>\n",
              "      <td>France</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541907</th>\n",
              "      <td>581587</td>\n",
              "      <td>23255</td>\n",
              "      <td>CHILDRENS CUTLERY CIRCUS PARADE</td>\n",
              "      <td>4</td>\n",
              "      <td>2011-12-09 12:50:00</td>\n",
              "      <td>4.15</td>\n",
              "      <td>12680</td>\n",
              "      <td>France</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541908</th>\n",
              "      <td>581587</td>\n",
              "      <td>22138</td>\n",
              "      <td>BAKING SET 9 PIECE RETROSPOT</td>\n",
              "      <td>3</td>\n",
              "      <td>2011-12-09 12:50:00</td>\n",
              "      <td>4.95</td>\n",
              "      <td>12680</td>\n",
              "      <td>France</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>541909 rows × 8 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-310e9e81-7f95-4aa3-9c43-ba8d24c198b8')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-92239daf-439b-4c59-b36a-2cf42ea57fb9\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-92239daf-439b-4c59-b36a-2cf42ea57fb9')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-92239daf-439b-4c59-b36a-2cf42ea57fb9 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-310e9e81-7f95-4aa3-9c43-ba8d24c198b8 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-310e9e81-7f95-4aa3-9c43-ba8d24c198b8');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 109
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Removendo informações incorretas da base:\n",
        "base_amostra.drop(base_amostra[(~base_amostra['InvoiceNo'].str.startswith('C'))\n",
        "& ((base_amostra['Quantity'] <= 0)\n",
        "| (base_amostra['UnitPrice'] <= 0))].index, inplace=True)"
      ],
      "metadata": {
        "id": "_y0vy7Jh5P7a"
      },
      "execution_count": 110,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### ***Features:***"
      ],
      "metadata": {
        "id": "9gPyNHKn7Yt9"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Tratamento de nulos e divergências:\n",
        "- Colunas `'CustomerID`' e `'StockCode`'\n",
        "---\n"
      ],
      "metadata": {
        "id": "c_zHj_Oy7HmI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# verificando valores ausentes:\n",
        "(amostra.isna().sum() / len(amostra)*100).sort_values(ascending=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rgfB1mp4CxS6",
        "outputId": "6b56ccf8-6ab5-4876-9b89-e8fab50fd4de"
      },
      "execution_count": 111,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Description    0.268311\n",
              "InvoiceNo      0.000000\n",
              "StockCode      0.000000\n",
              "Quantity       0.000000\n",
              "InvoiceDate    0.000000\n",
              "UnitPrice      0.000000\n",
              "CustomerID     0.000000\n",
              "Country        0.000000\n",
              "dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 111
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Após a remoção dos valores divergentes da base de dados, observamos uma redução significativa no percentual de dados nulos, com apenas `'Description'` apresentando **0.2%** de valores nulos. Com base nessa informação, podemos concluir que não é necessário realizar um tratamento adicional para lidar com valores nulos nesse momento, nos permitindo prosseguir com as análises e modelagem sem maiores intervenções nesse aspecto."
      ],
      "metadata": {
        "id": "N8lMuf2hD1zo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Percentual de valores 'nan' na coluna 'CustomerID' em relação a base:\n",
        "round(len(base_amostra[base_amostra['CustomerID'] == 'nan'])/ len(base_amostra)*100, 2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Sm26VDeYP2XW",
        "outputId": "d1abab39-fe3c-47cd-ee44-793b6d4060c8"
      },
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "24.58"
            ]
          },
          "metadata": {},
          "execution_count": 112
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Removendo os valores da base:\n",
        "base_amostra = base_amostra[base_amostra['CustomerID'] != 'nan']"
      ],
      "metadata": {
        "id": "mRTy0fBYLeEi"
      },
      "execution_count": 113,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Representando **24%** dos dados, removemos as informações 'nan' da coluna `'CustomerID'`."
      ],
      "metadata": {
        "id": "d5OMODGnNQu2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# verificando se há valores diferentes do padrão para 'InvoiceNo' (valores devem ter 5 dígitos)\n",
        "codigos_unicos_stock = base_amostra['StockCode'].tolist()\n",
        "codigos_invalidos = [codigo for codigo in codigos_unicos_stock if len(str(codigo)) != 5 or not str(codigo).isdigit()]\n",
        "f'Há {len(codigos_invalidos)} código(s) divergente(s). Alguns exemplos: {codigos_invalidos[:5]}'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "4dxb-785CJhE",
        "outputId": "19fba74c-eed0-461a-ef39-b6e6edf4f7b6"
      },
      "execution_count": 114,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "\"Há 35801 código(s) divergente(s). Alguns exemplos: ['85123A', '84406B', '84029G', '84029E', 'POST']\""
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "round(len(codigos_invalidos) / len(amostra)*100,2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EODWEh9oCSQF",
        "outputId": "7148b111-06a1-4b46-9947-8347b552b62b"
      },
      "execution_count": 115,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "6.61"
            ]
          },
          "metadata": {},
          "execution_count": 115
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Remover os valores invalidos da coluna 'StockCode':\n",
        "base_amostra = base_amostra[~base_amostra['StockCode'].isin(codigos_invalidos)]"
      ],
      "metadata": {
        "id": "5OUnE9FdUcU8"
      },
      "execution_count": 116,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Representando agora aproximandamente **7%** dos dados, removemos os valores divergentes da coluna `'StockCode'`."
      ],
      "metadata": {
        "id": "09FDDgTyEUr1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Filtro de data para recorte da base:\n",
        "- A base final será referente aos período de Outubro a Dezembro de 2011.\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "47IFe5Xu6d5Y"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Definir a data limite como dezembro de 2011:\n",
        "data_limite = pd.to_datetime('2011-12-01')\n",
        "\n",
        "# Filtrar os registros dos últimos 3 meses:\n",
        "base_amostra = base_amostra[base_amostra['InvoiceDate'] >= data_limite - pd.DateOffset(months=2)]\n",
        "\n",
        "# Exibir os registros filtrados:\n",
        "base_amostra['InvoiceDate'].dt.month.unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x_anPOR46d5Y",
        "outputId": "e8edf980-e8b2-48c0-959d-fe36e47c9fe4"
      },
      "execution_count": 117,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([10, 11, 12])"
            ]
          },
          "metadata": {},
          "execution_count": 117
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "base_amostra['InvoiceDate'].dt.year.unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lB0wZAIz8wtY",
        "outputId": "7e356c3b-0c28-4f60-9d78-6551b3d48155"
      },
      "execution_count": 118,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2011])"
            ]
          },
          "metadata": {},
          "execution_count": 118
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Proporção da base final em relação a base original\n",
        "round(len(base_amostra) / len(amostra)*100, 2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sZYMEcwB6d5Z",
        "outputId": "46f15535-5e0f-4400-bb02-8323b844fda3"
      },
      "execution_count": 119,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "22.92"
            ]
          },
          "metadata": {},
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Criando feature `'TotalPrice'`:\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "3HWZz4Aq4M-c"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Criando coluna com o cálculo do preço total da transação: 'TotalPrice':\n",
        "base_amostra['TotalPrice'] = base_amostra['Quantity'] * base_amostra['UnitPrice']\n",
        "base_amostra[['Quantity', 'UnitPrice', 'TotalPrice']]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "aNzQ5Z_eOabg",
        "outputId": "9d9fe9a0-1133-473f-c694-108f4a2f066e"
      },
      "execution_count": 120,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        Quantity  UnitPrice  TotalPrice\n",
              "370931        48       4.95      237.60\n",
              "370932        20       1.25       25.00\n",
              "370933         4       0.85        3.40\n",
              "370934        15       7.08      106.20\n",
              "370935         4       4.95       19.80\n",
              "...          ...        ...         ...\n",
              "541904        12       0.85       10.20\n",
              "541905         6       2.10       12.60\n",
              "541906         4       4.15       16.60\n",
              "541907         4       4.15       16.60\n",
              "541908         3       4.95       14.85\n",
              "\n",
              "[124193 rows x 3 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-89db1778-acc7-4c25-ad54-34f69c1264cc\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>Quantity</th>\n",
              "      <th>UnitPrice</th>\n",
              "      <th>TotalPrice</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>370931</th>\n",
              "      <td>48</td>\n",
              "      <td>4.95</td>\n",
              "      <td>237.60</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370932</th>\n",
              "      <td>20</td>\n",
              "      <td>1.25</td>\n",
              "      <td>25.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370933</th>\n",
              "      <td>4</td>\n",
              "      <td>0.85</td>\n",
              "      <td>3.40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370934</th>\n",
              "      <td>15</td>\n",
              "      <td>7.08</td>\n",
              "      <td>106.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370935</th>\n",
              "      <td>4</td>\n",
              "      <td>4.95</td>\n",
              "      <td>19.80</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541904</th>\n",
              "      <td>12</td>\n",
              "      <td>0.85</td>\n",
              "      <td>10.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541905</th>\n",
              "      <td>6</td>\n",
              "      <td>2.10</td>\n",
              "      <td>12.60</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541906</th>\n",
              "      <td>4</td>\n",
              "      <td>4.15</td>\n",
              "      <td>16.60</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541907</th>\n",
              "      <td>4</td>\n",
              "      <td>4.15</td>\n",
              "      <td>16.60</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541908</th>\n",
              "      <td>3</td>\n",
              "      <td>4.95</td>\n",
              "      <td>14.85</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>124193 rows × 3 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-89db1778-acc7-4c25-ad54-34f69c1264cc')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-14b48377-552a-499b-8cde-b5f029947944\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-14b48377-552a-499b-8cde-b5f029947944')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-14b48377-552a-499b-8cde-b5f029947944 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-89db1778-acc7-4c25-ad54-34f69c1264cc button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-89db1778-acc7-4c25-ad54-34f69c1264cc');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Verificandos e tratando informações estatísticas sobre as variáveis `'TotalPrice'`, `'UnitPrice'` e `'Quantity'`:\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "LkbJy8GZcAwA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "base_amostra[['TotalPrice', 'UnitPrice', 'Quantity']].describe().T"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "6-fy8qZHfRYX",
        "outputId": "547cae3b-62ab-4a91-a642-81dc83927847"
      },
      "execution_count": 121,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               count       mean         std        min   25%   50%    75%  \\\n",
              "TotalPrice  124193.0  18.167294  679.044724 -168469.60  3.40  9.95  17.70   \n",
              "UnitPrice   124193.0   2.763063    3.129232       0.04  1.25  1.79   3.75   \n",
              "Quantity    124193.0  10.716425  327.375629  -80995.00  2.00  4.00  12.00   \n",
              "\n",
              "                 max  \n",
              "TotalPrice  168469.6  \n",
              "UnitPrice      165.0  \n",
              "Quantity     80995.0  "
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-73ade40d-33d0-42fc-bfc5-4b21ea68fde9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>count</th>\n",
              "      <th>mean</th>\n",
              "      <th>std</th>\n",
              "      <th>min</th>\n",
              "      <th>25%</th>\n",
              "      <th>50%</th>\n",
              "      <th>75%</th>\n",
              "      <th>max</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>TotalPrice</th>\n",
              "      <td>124193.0</td>\n",
              "      <td>18.167294</td>\n",
              "      <td>679.044724</td>\n",
              "      <td>-168469.60</td>\n",
              "      <td>3.40</td>\n",
              "      <td>9.95</td>\n",
              "      <td>17.70</td>\n",
              "      <td>168469.6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>UnitPrice</th>\n",
              "      <td>124193.0</td>\n",
              "      <td>2.763063</td>\n",
              "      <td>3.129232</td>\n",
              "      <td>0.04</td>\n",
              "      <td>1.25</td>\n",
              "      <td>1.79</td>\n",
              "      <td>3.75</td>\n",
              "      <td>165.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Quantity</th>\n",
              "      <td>124193.0</td>\n",
              "      <td>10.716425</td>\n",
              "      <td>327.375629</td>\n",
              "      <td>-80995.00</td>\n",
              "      <td>2.00</td>\n",
              "      <td>4.00</td>\n",
              "      <td>12.00</td>\n",
              "      <td>80995.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-73ade40d-33d0-42fc-bfc5-4b21ea68fde9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-133234e0-b536-4a70-ae7b-c81430e8cff6\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-133234e0-b536-4a70-ae7b-c81430e8cff6')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-133234e0-b536-4a70-ae7b-c81430e8cff6 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-73ade40d-33d0-42fc-bfc5-4b21ea68fde9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-73ade40d-33d0-42fc-bfc5-4b21ea68fde9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 121
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "colors = ['r', 'g', 'b']\n",
        "base_amostra[['TotalPrice', 'UnitPrice', 'Quantity']].plot(kind='box', vert=False)\n",
        "\n",
        "outliers = base_amostra[['TotalPrice', 'UnitPrice', 'Quantity']].describe().loc[['min', 'max']].values.round()\n",
        "\n",
        "for i, limit in enumerate(outliers.T):\n",
        "    color = colors[i % len(colors)]\n",
        "\n",
        "    plt.axvline(x=limit[0], color=color, linestyle='--')\n",
        "    plt.axvline(x=limit[1], color=color, linestyle='--')\n",
        "\n",
        "    for j, value in enumerate(limit):\n",
        "        plt.text(value, i+1.1, f'{value}', ha='center', va='bottom')\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 434
        },
        "id": "68tWgTRLgh4e",
        "outputId": "324aac96-339d-47b7-8c4c-6e37ef61884b"
      },
      "execution_count": 122,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "outliers_df = pd.DataFrame(outliers, columns=['TotalPrice', 'UnitPrice', 'Quantity'], index=['min', 'max'])\n",
        "outliers_df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        },
        "id": "xw7IIlpqoa2Q",
        "outputId": "b4e177e1-3c4d-4ce6-b815-a94eafa495bf"
      },
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     TotalPrice  UnitPrice  Quantity\n",
              "min   -168470.0        0.0  -80995.0\n",
              "max    168470.0      165.0   80995.0"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-4e1d5847-0314-4584-a2fc-c380d63af695\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>TotalPrice</th>\n",
              "      <th>UnitPrice</th>\n",
              "      <th>Quantity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-168470.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>-80995.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>168470.0</td>\n",
              "      <td>165.0</td>\n",
              "      <td>80995.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4e1d5847-0314-4584-a2fc-c380d63af695')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-6b96a444-a8f2-459b-bae0-b86ed3f5eec0\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-6b96a444-a8f2-459b-bae0-b86ed3f5eec0')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-6b96a444-a8f2-459b-bae0-b86ed3f5eec0 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-4e1d5847-0314-4584-a2fc-c380d63af695 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-4e1d5847-0314-4584-a2fc-c380d63af695');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 123
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As colunas `'TotalPrice'` e `'Quantity'` exibem valores discrepantes, resultando em um **alto desvio padrão** de 679 e 327, respectivamente. A fim de melhor equilibrar a distribuição dos dados, será aplicada a estratégia de remover esses valores discrepantes, também conhecidos como outliers, em ambas as colunas."
      ],
      "metadata": {
        "id": "4YfDu1W--_8x"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Removendo valores max e mim de 'TotalPrice' e 'Quantity':\n",
        "base_amostra = base_amostra[(base_amostra['TotalPrice'] != base_amostra['TotalPrice'].max()) & (base_amostra['Quantity'] != base_amostra['Quantity'].min())]"
      ],
      "metadata": {
        "id": "XHLF5Bpk_i_r"
      },
      "execution_count": 124,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "base_amostra[['TotalPrice', 'UnitPrice', 'Quantity']].describe().T"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "v6mdERFJGuLc",
        "outputId": "1dea868c-1253-451c-b070-1606d02ccc71"
      },
      "execution_count": 125,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               count       mean        std      min   25%   50%    75%     max\n",
              "TotalPrice  124191.0  18.167586  63.523387 -1836.00  3.40  9.95  17.70  4992.0\n",
              "UnitPrice   124191.0   2.763074   3.129256     0.04  1.25  1.79   3.75   165.0\n",
              "Quantity    124191.0  10.716598  39.101019 -1296.00  2.00  4.00  12.00  4800.0"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-66794a67-5230-476e-a113-2216757eacda\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>count</th>\n",
              "      <th>mean</th>\n",
              "      <th>std</th>\n",
              "      <th>min</th>\n",
              "      <th>25%</th>\n",
              "      <th>50%</th>\n",
              "      <th>75%</th>\n",
              "      <th>max</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>TotalPrice</th>\n",
              "      <td>124191.0</td>\n",
              "      <td>18.167586</td>\n",
              "      <td>63.523387</td>\n",
              "      <td>-1836.00</td>\n",
              "      <td>3.40</td>\n",
              "      <td>9.95</td>\n",
              "      <td>17.70</td>\n",
              "      <td>4992.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>UnitPrice</th>\n",
              "      <td>124191.0</td>\n",
              "      <td>2.763074</td>\n",
              "      <td>3.129256</td>\n",
              "      <td>0.04</td>\n",
              "      <td>1.25</td>\n",
              "      <td>1.79</td>\n",
              "      <td>3.75</td>\n",
              "      <td>165.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Quantity</th>\n",
              "      <td>124191.0</td>\n",
              "      <td>10.716598</td>\n",
              "      <td>39.101019</td>\n",
              "      <td>-1296.00</td>\n",
              "      <td>2.00</td>\n",
              "      <td>4.00</td>\n",
              "      <td>12.00</td>\n",
              "      <td>4800.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-66794a67-5230-476e-a113-2216757eacda')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-3184ef97-1d4e-4603-8e22-3b133a81b24c\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-3184ef97-1d4e-4603-8e22-3b133a81b24c')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-3184ef97-1d4e-4603-8e22-3b133a81b24c button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-66794a67-5230-476e-a113-2216757eacda button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-66794a67-5230-476e-a113-2216757eacda');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 125
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Observamos uma diminuição significativa nos desvios padrão de `'TotalPrice'` e `'Quantity'`, para 63 e 39, respectivamente. Embora ainda sejam valores relativamente elevados, percebe-se uma redução substancial em relação aos números anteriores."
      ],
      "metadata": {
        "id": "FKPZsMg9HAJp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "colors = ['r', 'g', 'b']\n",
        "base_amostra[['TotalPrice', 'UnitPrice', 'Quantity']].plot(kind='box', vert=False)\n",
        "\n",
        "outliers = base_amostra[['TotalPrice', 'UnitPrice', 'Quantity']].describe().loc[['min', 'max']].values.round()\n",
        "\n",
        "for i, limit in enumerate(outliers.T):\n",
        "    color = colors[i % len(colors)]\n",
        "\n",
        "    plt.axvline(x=limit[0], color=color, linestyle='--')\n",
        "    plt.axvline(x=limit[1], color=color, linestyle='--')\n",
        "\n",
        "    for j, value in enumerate(limit):\n",
        "        plt.text(value, i+1.1, f'{value}', ha='center', va='bottom')\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 434
        },
        "id": "erzJL4yYGoqp",
        "outputId": "1fd6e2ff-56d3-4f89-d073-88ff3baae815"
      },
      "execution_count": 126,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### ***Target:***"
      ],
      "metadata": {
        "id": "nLGAoqmw7uUY"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Tratamento, criação e análise da target:\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "xMVgGxZd4XwR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Valores únicos em 'InvoiceNo':\n",
        "base_amostra['InvoiceNo'] = base_amostra['InvoiceNo'].astype(str)\n",
        "base_amostra[base_amostra['InvoiceNo'].str.contains('C')]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "w8p0AQrg3zmp",
        "outputId": "7db6b685-d524-4741-eae1-4cdcd5855132"
      },
      "execution_count": 127,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       InvoiceNo StockCode                         Description  Quantity  \\\n",
              "372165   C569233     22350               ILLUSTRATED CAT BOWL         -1   \n",
              "372166   C569233     22467                   GUMBALL COAT RACK        -2   \n",
              "372881   C569251     85066         CREAM SWEETHEART MINI CHEST        -1   \n",
              "373247   C569318     22628    PICNIC BOXES SET OF 3 RETROSPOT         -1   \n",
              "373248   C569318     22328  ROUND SNACK BOXES SET OF 4 FRUITS         -3   \n",
              "...          ...       ...                                 ...       ...   \n",
              "540448   C581490     22178     VICTORIAN GLASS HANGING T-LIGHT       -12   \n",
              "540449   C581490     23144     ZINC T-LIGHT HOLDER STARS SMALL       -11   \n",
              "541715   C581568     21258          VICTORIAN SEWING BOX LARGE        -5   \n",
              "541716   C581569     84978    HANGING HEART JAR T-LIGHT HOLDER        -1   \n",
              "541717   C581569     20979       36 PENCILS TUBE RED RETROSPOT        -5   \n",
              "\n",
              "               InvoiceDate  UnitPrice CustomerID         Country  TotalPrice  \n",
              "372165 2011-10-02 15:26:00       2.55      17841  United Kingdom       -2.55  \n",
              "372166 2011-10-02 15:26:00       2.55      17841  United Kingdom       -5.10  \n",
              "372881 2011-10-03 10:37:00      12.75      12957  United Kingdom      -12.75  \n",
              "373247 2011-10-03 12:14:00       4.95      13505     Switzerland       -4.95  \n",
              "373248 2011-10-03 12:14:00       2.95      13505     Switzerland       -8.85  \n",
              "...                    ...        ...        ...             ...         ...  \n",
              "540448 2011-12-09 09:57:00       1.95      14397  United Kingdom      -23.40  \n",
              "540449 2011-12-09 09:57:00       0.83      14397  United Kingdom       -9.13  \n",
              "541715 2011-12-09 11:57:00      10.95      15311  United Kingdom      -54.75  \n",
              "541716 2011-12-09 11:58:00       1.25      17315  United Kingdom       -1.25  \n",
              "541717 2011-12-09 11:58:00       1.25      17315  United Kingdom       -6.25  \n",
              "\n",
              "[2275 rows x 9 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-7a965745-c8c3-41e0-9cda-40d0e3bc16e0\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>InvoiceNo</th>\n",
              "      <th>StockCode</th>\n",
              "      <th>Description</th>\n",
              "      <th>Quantity</th>\n",
              "      <th>InvoiceDate</th>\n",
              "      <th>UnitPrice</th>\n",
              "      <th>CustomerID</th>\n",
              "      <th>Country</th>\n",
              "      <th>TotalPrice</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>372165</th>\n",
              "      <td>C569233</td>\n",
              "      <td>22350</td>\n",
              "      <td>ILLUSTRATED CAT BOWL</td>\n",
              "      <td>-1</td>\n",
              "      <td>2011-10-02 15:26:00</td>\n",
              "      <td>2.55</td>\n",
              "      <td>17841</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-2.55</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>372166</th>\n",
              "      <td>C569233</td>\n",
              "      <td>22467</td>\n",
              "      <td>GUMBALL COAT RACK</td>\n",
              "      <td>-2</td>\n",
              "      <td>2011-10-02 15:26:00</td>\n",
              "      <td>2.55</td>\n",
              "      <td>17841</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-5.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>372881</th>\n",
              "      <td>C569251</td>\n",
              "      <td>85066</td>\n",
              "      <td>CREAM SWEETHEART MINI CHEST</td>\n",
              "      <td>-1</td>\n",
              "      <td>2011-10-03 10:37:00</td>\n",
              "      <td>12.75</td>\n",
              "      <td>12957</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-12.75</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>373247</th>\n",
              "      <td>C569318</td>\n",
              "      <td>22628</td>\n",
              "      <td>PICNIC BOXES SET OF 3 RETROSPOT</td>\n",
              "      <td>-1</td>\n",
              "      <td>2011-10-03 12:14:00</td>\n",
              "      <td>4.95</td>\n",
              "      <td>13505</td>\n",
              "      <td>Switzerland</td>\n",
              "      <td>-4.95</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>373248</th>\n",
              "      <td>C569318</td>\n",
              "      <td>22328</td>\n",
              "      <td>ROUND SNACK BOXES SET OF 4 FRUITS</td>\n",
              "      <td>-3</td>\n",
              "      <td>2011-10-03 12:14:00</td>\n",
              "      <td>2.95</td>\n",
              "      <td>13505</td>\n",
              "      <td>Switzerland</td>\n",
              "      <td>-8.85</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>540448</th>\n",
              "      <td>C581490</td>\n",
              "      <td>22178</td>\n",
              "      <td>VICTORIAN GLASS HANGING T-LIGHT</td>\n",
              "      <td>-12</td>\n",
              "      <td>2011-12-09 09:57:00</td>\n",
              "      <td>1.95</td>\n",
              "      <td>14397</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-23.40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>540449</th>\n",
              "      <td>C581490</td>\n",
              "      <td>23144</td>\n",
              "      <td>ZINC T-LIGHT HOLDER STARS SMALL</td>\n",
              "      <td>-11</td>\n",
              "      <td>2011-12-09 09:57:00</td>\n",
              "      <td>0.83</td>\n",
              "      <td>14397</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-9.13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541715</th>\n",
              "      <td>C581568</td>\n",
              "      <td>21258</td>\n",
              "      <td>VICTORIAN SEWING BOX LARGE</td>\n",
              "      <td>-5</td>\n",
              "      <td>2011-12-09 11:57:00</td>\n",
              "      <td>10.95</td>\n",
              "      <td>15311</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-54.75</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541716</th>\n",
              "      <td>C581569</td>\n",
              "      <td>84978</td>\n",
              "      <td>HANGING HEART JAR T-LIGHT HOLDER</td>\n",
              "      <td>-1</td>\n",
              "      <td>2011-12-09 11:58:00</td>\n",
              "      <td>1.25</td>\n",
              "      <td>17315</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-1.25</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541717</th>\n",
              "      <td>C581569</td>\n",
              "      <td>20979</td>\n",
              "      <td>36 PENCILS TUBE RED RETROSPOT</td>\n",
              "      <td>-5</td>\n",
              "      <td>2011-12-09 11:58:00</td>\n",
              "      <td>1.25</td>\n",
              "      <td>17315</td>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>-6.25</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2275 rows × 9 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7a965745-c8c3-41e0-9cda-40d0e3bc16e0')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-d59879d8-ad31-44bb-85cd-c4a2527a55f2\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-d59879d8-ad31-44bb-85cd-c4a2527a55f2')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-d59879d8-ad31-44bb-85cd-c4a2527a55f2 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-7a965745-c8c3-41e0-9cda-40d0e3bc16e0 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-7a965745-c8c3-41e0-9cda-40d0e3bc16e0');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 127
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Criando coluna 'Status' para pontuar os cancelamentos:\n",
        "base_amostra['Status'] = base_amostra['InvoiceNo'].apply(lambda x: 'Cancelado' if str(x).startswith('C') else 'Não Cancelado')\n",
        "base_amostra[['Status', 'InvoiceNo']]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "j6k351phNM8W",
        "outputId": "3dcbe51a-7f4b-4c80-b24b-e959ae627846"
      },
      "execution_count": 128,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               Status InvoiceNo\n",
              "370931  Não Cancelado    569203\n",
              "370932  Não Cancelado    569203\n",
              "370933  Não Cancelado    569204\n",
              "370934  Não Cancelado    569204\n",
              "370935  Não Cancelado    569204\n",
              "...               ...       ...\n",
              "541904  Não Cancelado    581587\n",
              "541905  Não Cancelado    581587\n",
              "541906  Não Cancelado    581587\n",
              "541907  Não Cancelado    581587\n",
              "541908  Não Cancelado    581587\n",
              "\n",
              "[124191 rows x 2 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-44fa72a4-7410-4ed2-8ce8-8bd95a303e8f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>Status</th>\n",
              "      <th>InvoiceNo</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>370931</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>569203</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370932</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>569203</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370933</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>569204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370934</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>569204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370935</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>569204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541904</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>581587</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541905</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>581587</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541906</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>581587</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541907</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>581587</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>541908</th>\n",
              "      <td>Não Cancelado</td>\n",
              "      <td>581587</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>124191 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-44fa72a4-7410-4ed2-8ce8-8bd95a303e8f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-f67b988f-0ef3-476d-aa5c-9f32301e78b9\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-f67b988f-0ef3-476d-aa5c-9f32301e78b9')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-f67b988f-0ef3-476d-aa5c-9f32301e78b9 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-44fa72a4-7410-4ed2-8ce8-8bd95a303e8f button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-44fa72a4-7410-4ed2-8ce8-8bd95a303e8f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 128
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Proporção da target na base:\n",
        "colors = ['#309b91', '#054863']\n",
        "base_amostra['Status'].value_counts().plot(kind = 'bar', color=colors)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 553
        },
        "id": "fvD3Rky6fNCa",
        "outputId": "331a9053-c5e1-4670-f07c-a16157d76416"
      },
      "execution_count": 129,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 129
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "round(base_amostra['Status'].value_counts() / len(base_amostra)*100, 2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X6njAHywmV2M",
        "outputId": "35858add-f4fd-4612-e955-e15ed2f147b6"
      },
      "execution_count": 130,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Não Cancelado    98.17\n",
              "Cancelado         1.83\n",
              "Name: Status, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 130
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Verificamos que o percentual de compras canceladas é bastante baixo, representando menos de **2%** do total de registros. Para lidar com esse desequilíbrio nos dados, adotaremos no momento da modelagem a técnica de ***Under-sampling.*** Essa abordagem envolve extrair um subconjunto aleatório da classe majoritária, preservando as características da classe minoritária, sendo ideal para situações onde você tem grandes volumes de dados.\n",
        "\n",
        "- O objetivo é **aumentar a representatividade da classe minoritária na base de dados, proporcionando um melhor desempenho do modelo e evitando possíveis viés decorrentes do desequilíbrio entre as classes.** Dessa forma, garantimos que o modelo seja capaz de capturar adequadamente os padrões e características da classe cancelada, contribuindo para uma melhor tomada de decisão."
      ],
      "metadata": {
        "id": "1Gk294Ws60lO"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### **EDA** sobre as informações base:"
      ],
      "metadata": {
        "id": "AmoBodxzZepr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Resumo das análises:\n",
        "\n",
        "- Análise Regional (pais e continente).\n",
        "- Análise dos itens e suas possíveis categorias.\n",
        "- Análise temporal (vendas e cancelamentos em intervalos de hora).\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "On7WFuG0fWOT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Compras por pais:\n",
        "base_amostra['Country'].value_counts() / len(amostra)*100"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ego2o_BxANBo",
        "outputId": "e06c1c43-9e6e-4e32-b98a-0555ddf3c650"
      },
      "execution_count": 131,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "United Kingdom     20.633538\n",
              "France              0.470928\n",
              "Germany             0.468344\n",
              "EIRE                0.384935\n",
              "Spain               0.118470\n",
              "Portugal            0.111642\n",
              "Belgium             0.103338\n",
              "Netherlands         0.102969\n",
              "Switzerland         0.088945\n",
              "Italy               0.062741\n",
              "Norway              0.062372\n",
              "Cyprus              0.046871\n",
              "USA                 0.046318\n",
              "Channel Islands     0.031740\n",
              "Finland             0.030079\n",
              "Australia           0.027495\n",
              "Austria             0.022144\n",
              "Sweden              0.021590\n",
              "Denmark             0.018453\n",
              "Poland              0.012917\n",
              "Japan               0.010334\n",
              "RSA                 0.009965\n",
              "Iceland             0.009965\n",
              "Greece              0.006459\n",
              "Israel              0.005167\n",
              "Singapore           0.004982\n",
              "Malta               0.002768\n",
              "Czech Republic      0.001845\n",
              "Name: Country, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 131
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Média do 'TotalPrice' por pais:\n",
        "base_amostra.groupby('Country')['TotalPrice'].mean().sort_values(ascending = False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VulUor0deerg",
        "outputId": "54aa894a-ac13-45c3-e5e8-9ff69b4a0349"
      },
      "execution_count": 132,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Country\n",
              "Japan              183.771250\n",
              "Australia          158.874832\n",
              "Netherlands        135.362778\n",
              "Sweden              67.646410\n",
              "Singapore           34.301481\n",
              "Norway              31.973521\n",
              "Denmark             31.515200\n",
              "Switzerland         27.300996\n",
              "EIRE                25.829727\n",
              "Iceland             24.895185\n",
              "Channel Islands     22.429302\n",
              "Germany             21.874275\n",
              "France              21.632610\n",
              "Czech Republic      21.573000\n",
              "Belgium             20.683107\n",
              "Finland             19.638160\n",
              "Italy               19.310382\n",
              "Poland              18.348143\n",
              "Austria             18.190500\n",
              "RSA                 17.584444\n",
              "United Kingdom      16.902602\n",
              "Cyprus              16.711142\n",
              "Greece              15.599714\n",
              "Portugal            15.486727\n",
              "Spain               14.598723\n",
              "Malta               12.720667\n",
              "USA                  4.837331\n",
              "Israel               3.802500\n",
              "Name: TotalPrice, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 132
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Criando coluna de 'Continente':\n",
        "dict_continent = {\n",
        "    'United Kingdom': 'Europa',\n",
        "    'Switzerland': 'Europa',\n",
        "    'France': 'Europa',\n",
        "    'Spain': 'Europa',\n",
        "    'Germany': 'Europa',\n",
        "    'Belgium': 'Europa',\n",
        "    'Israel': 'Ásia',\n",
        "    'Channel Islands': 'Europa',\n",
        "    'Australia': 'Oceania',\n",
        "    'Netherlands': 'Europa',\n",
        "    'EIRE': 'Europa',\n",
        "    'Portugal': 'Europa',\n",
        "    'Italy': 'Europa',\n",
        "    'Denmark': 'Europa',\n",
        "    'Cyprus': 'Europa',\n",
        "    'Czech Republic': 'Europa',\n",
        "    'USA': 'América do Norte',\n",
        "    'Norway': 'Europa',\n",
        "    'Finland': 'Europa',\n",
        "    'RSA': 'África',\n",
        "    'Singapore': 'Ásia',\n",
        "    'Sweden': 'Europa',\n",
        "    'Austria': 'Europa',\n",
        "    'Japan': 'Ásia',\n",
        "    'Iceland': 'Europa',\n",
        "    'Poland': 'Europa',\n",
        "    'Malta': 'Europa',\n",
        "    'Greece': 'Europa'\n",
        "}\n",
        "\n",
        "base_amostra['Continent'] = base_amostra['Country'].map(dict_continent)"
      ],
      "metadata": {
        "id": "F3AlcqKVa-ZO"
      },
      "execution_count": 133,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Média de 'UnitPrice' por continente:\n",
        "base_amostra.groupby('Continent')['UnitPrice'].mean().sort_values(ascending = False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IqcqXC-FcXqj",
        "outputId": "74c67bcf-1ded-4a35-f550-9ed3b82e525d"
      },
      "execution_count": 134,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Continent\n",
              "África              4.418889\n",
              "Ásia                3.343514\n",
              "Oceania             3.012416\n",
              "Europa              2.762968\n",
              "América do Norte    2.054502\n",
              "Name: UnitPrice, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 134
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Média de 'UnitPrice' por continente:\n",
        "base_amostra.groupby('Continent')['TotalPrice'].mean().sort_values(ascending = False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LpYD7LCqdGL9",
        "outputId": "c00af24d-8dd2-466c-bf4b-c70fa2db1cf1"
      },
      "execution_count": 135,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Continent\n",
              "Oceania             158.874832\n",
              "Ásia                102.016216\n",
              "Europa               17.950034\n",
              "África               17.584444\n",
              "América do Norte      4.837331\n",
              "Name: TotalPrice, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 135
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Média de 'Country' por continente:\n",
        "base_amostra.groupby('Country')['UnitPrice'].mean().sort_values(ascending = False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yzfWSIKoZj7L",
        "outputId": "2a16f45e-f4c6-48a9-f456-ced640b8f165"
      },
      "execution_count": 136,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Country\n",
              "Singapore          5.642222\n",
              "Cyprus             5.013622\n",
              "Malta              4.688667\n",
              "RSA                4.418889\n",
              "Italy              4.025735\n",
              "Finland            3.590798\n",
              "Channel Islands    3.437791\n",
              "Greece             3.381714\n",
              "EIRE               3.299056\n",
              "Belgium            3.284161\n",
              "Norway             3.264556\n",
              "Spain              3.192757\n",
              "Australia          3.012416\n",
              "Austria            2.852667\n",
              "Germany            2.793775\n",
              "United Kingdom     2.740471\n",
              "Japan              2.709464\n",
              "Portugal           2.701983\n",
              "Poland             2.698143\n",
              "Switzerland        2.676286\n",
              "France             2.675513\n",
              "Netherlands        2.410287\n",
              "Israel             2.395000\n",
              "Iceland            2.283333\n",
              "USA                2.054502\n",
              "Sweden             1.936239\n",
              "Denmark            1.873700\n",
              "Czech Republic     1.083000\n",
              "Name: UnitPrice, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 136
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Conforme mencionado anteriormente, a maioria das compras foi realizada no Reino Unido (Unit Kingdom), porém o país ocupa a 21ª posição em termos de média total de compras. Por outro lado, o Japão possui a maior média total de compras, apesar de não representar nem 1% das vendas totais.\n",
        "\n",
        "- No que diz respeito aos continentes, constata-se que o continente africano apresenta os itens mais caros, em média, enquanto ocupa a segunda posição em relação ao menor valor total de compras. Já a Oceania é o continente onde as pessoas gastaram mais em média."
      ],
      "metadata": {
        "id": "uLysCn7kY6w4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Qtd de itens vendidos:\n",
        "base_amostra.groupby(['StockCode', 'Description'])['StockCode'].count().sort_values(ascending=False).head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IRSS70bpi5uz",
        "outputId": "e7235dba-ee44-4c15-8a86-19b15588ca5d"
      },
      "execution_count": 137,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "StockCode  Description                       \n",
              "22086      PAPER CHAIN KIT 50'S CHRISTMAS        673\n",
              "23084      RABBIT NIGHT LIGHT                    655\n",
              "23355      HOT WATER BOTTLE KEEP CALM            538\n",
              "22910      PAPER CHAIN KIT VINTAGE CHRISTMAS     465\n",
              "23344      JUMBO BAG 50'S CHRISTMAS              435\n",
              "22197      POPCORN HOLDER                        408\n",
              "22865      HAND WARMER OWL DESIGN                406\n",
              "22423      REGENCY CAKESTAND 3 TIER              399\n",
              "84879      ASSORTED COLOUR BIRD ORNAMENT         398\n",
              "22578      WOODEN STAR CHRISTMAS SCANDINAVIAN    380\n",
              "Name: StockCode, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 137
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Na base de dados, não há uma segmentação de itens por categoria. Ao analisar os 10 itens mais vendidos, observamos, por exemplo, que eles consistem em itens de decoração e utensílios gerais. Dentro da categoria de decoração, identificamos 3 itens com tema natalino, o que faz sentido, uma vez que estamos verificando os últimos 3 meses do ano.\n",
        "\n",
        "- No entanto, essa análise superficial pode não ser precisa o suficiente para compreender completamente as informações. Uma alternativa viável seria utilizar um modelo de clusterização, como o KNN ou um SVM, para segmentar de forma mais precisa e obter uma compreensão mais detalhada das categorias presentes. No entanto, essa abordagem requereria uma categorização manual inicial, a fim de obter dados reais para treinar e testar o modelo. Considerando que já possuímos a descrição do item e que a coluna de categorias não alteraria significativamente a capacidade de predição do modelo, optamos por não realizar essa etapa."
      ],
      "metadata": {
        "id": "FYg_WXsYmZ3h"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Tratamento de data e hora:\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "SacIufykcwos"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Criando coluna com intervalos de horas:\n",
        "base_amostra['InvoiceHour'] = base_amostra['InvoiceDate'].dt.strftime('%H:%M:%S')\n",
        "intervalos = pd.interval_range(start=pd.to_timedelta('00:00:00'), end=pd.to_timedelta('23:00:00'), freq='1H')\n",
        "\n",
        "base_amostra['InvoiceHour_Intervalo'] = pd.cut(pd.to_timedelta(base_amostra['InvoiceHour']), bins=intervalos, labels=intervalos[:-1])\n",
        "\n",
        "base_amostra['InvoiceHour'] = base_amostra['InvoiceHour_Intervalo'].apply(lambda x: datetime.strftime(datetime(1900, 1, 1) + x.left, '%H'f'h'))"
      ],
      "metadata": {
        "id": "4XTTdeiZrrRr"
      },
      "execution_count": 138,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Vendas 'não canceladas' por horário:\n",
        "invoice_hour_counts_not = base_amostra[~base_amostra['InvoiceNo'].str.startswith('C')].groupby('InvoiceHour')['InvoiceNo'].count()\n",
        "\n",
        "plt.figure(figsize=(10, 6))\n",
        "sns.barplot(x=invoice_hour_counts_not.index, y=invoice_hour_counts_not.values)\n",
        "plt.xlabel('Hora do Pedido')\n",
        "plt.ylabel('Contagem')\n",
        "plt.title('Contagem de Pedidos por Hora')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "id": "wNNk0t0RuCe1",
        "outputId": "7a079b77-a7de-4300-95d1-3c52ca2b37c1"
      },
      "execution_count": 139,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# itens não cancelados no período das 12h\n",
        "base_amostra[(~base_amostra['InvoiceNo'].str.startswith('C')) & (base_amostra['InvoiceHour'] == '12h')].groupby('Description')['Description'].count().sort_values(ascending=False).head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XPrjc1fxfSGW",
        "outputId": "c3e768f9-90b4-4728-fda0-e3fd418dd307"
      },
      "execution_count": 140,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Description\n",
              "PAPER CHAIN KIT 50'S CHRISTMAS        126\n",
              "HOT WATER BOTTLE KEEP CALM            111\n",
              "RABBIT NIGHT LIGHT                    106\n",
              "JUMBO BAG 50'S CHRISTMAS               94\n",
              "REX CASH+CARRY JUMBO SHOPPER           93\n",
              "PAPER CHAIN KIT VINTAGE CHRISTMAS      83\n",
              "GARDENERS KNEELING PAD KEEP CALM       81\n",
              "POPCORN HOLDER                         78\n",
              "HAND WARMER OWL DESIGN                 77\n",
              "WOODEN STAR CHRISTMAS SCANDINAVIAN     76\n",
              "Name: Description, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 140
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Vendas canceladas por horário:\n",
        "invoice_hour_counts = base_amostra[base_amostra['InvoiceNo'].str.startswith('C')].groupby('InvoiceHour')['InvoiceNo'].count()\n",
        "\n",
        "plt.figure(figsize=(10, 6))\n",
        "sns.barplot(x=invoice_hour_counts.index, y=invoice_hour_counts.values)\n",
        "plt.xlabel('Hora do Pedido')\n",
        "plt.ylabel('Contagem')\n",
        "plt.title('Contagem de Pedidos Cancelados por Hora')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "id": "NOvm7PiiuPex",
        "outputId": "97548adb-1f5d-42f5-8823-6da2b81ba9aa"
      },
      "execution_count": 141,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- A maioria das vendas ocorre entre as 11h da manhã e as 14h, com pico ao 12h indicando um período de maior atividade comercial.\n",
        "\n",
        "- Além disso, observa-se que os cancelamentos também apresentam um volume significativo durante esse mesmo período, com pico as 11h, indicando uma possível relação entre as horas de maior volume de vendas e os cancelamentos. Outro pico de cancelamentos ocorre às 16h da tarde."
      ],
      "metadata": {
        "id": "0jMmCOXWvvnR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# itens cancelados no período das 11h\n",
        "base_amostra[(base_amostra['InvoiceNo'].str.startswith('C')) & (base_amostra['InvoiceHour'] == '11h')].groupby('Description')['Description'].count().sort_values(ascending=False).head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JNMiEOc8deQo",
        "outputId": "c38ed37e-e83f-4538-88ed-2f809d8a094d"
      },
      "execution_count": 142,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Description\n",
              "CHILDRENS APRON SPACEBOY DESIGN        3\n",
              "FELTCRAFT CUSHION OWL                  3\n",
              "LUNCH BAG CARS BLUE                    3\n",
              "JUMBO BAG 50'S CHRISTMAS               3\n",
              "JUMBO STORAGE BAG SUKI                 3\n",
              "JUMBO BAG PINK POLKADOT                3\n",
              "POPCORN HOLDER                         3\n",
              "SPACEBOY LUNCH BOX                     2\n",
              "SET OF 60 I LOVE LONDON CAKE CASES     2\n",
              "SET OF 4 KNICK KNACK TINS LONDON       2\n",
              "Name: Description, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 142
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# itens cancelados no período das 16h\n",
        "base_amostra[(base_amostra['InvoiceNo'].str.startswith('C')) & (base_amostra['InvoiceHour'] == '16h')].groupby('Description')['Description'].count().sort_values(ascending=False).head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IscVzPYCWd-l",
        "outputId": "8198c3ae-f430-45cb-fa08-224c7c8c63f0"
      },
      "execution_count": 143,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Description\n",
              "PARISIENNE JEWELLERY DRAWER            4\n",
              "SET OF 3 REGENCY CAKE TINS             3\n",
              "SET OF 3 CAKE TINS PANTRY DESIGN       3\n",
              "CLASSIC GLASS COOKIE JAR               3\n",
              "SET OF TEA COFFEE SUGAR TINS PANTRY    3\n",
              "VINTAGE DOILY DELUXE SEWING KIT        3\n",
              "COLOUR GLASS. STAR T-LIGHT HOLDER      3\n",
              "REGENCY CAKESTAND 3 TIER               3\n",
              "ROSES REGENCY TEACUP AND SAUCER        2\n",
              "REGENCY MILK JUG PINK                  2\n",
              "Name: Description, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 143
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Durante os períodos de maior número de cancelamentos, não observamos uma concentração significativa em nenhum item específico, o que sugere que os cancelamentos ocorrem de forma aleatória."
      ],
      "metadata": {
        "id": "tlSYtNSaegNS"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### ***Aplicação do Modelo:***\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "sjOPGz8B8AJK"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Resumos dos passos:\n",
        "\n",
        "- Separação das features numéricas e categóricas.\n",
        "- Criação da pipeline para tratamento das features.\n",
        "  - Método RobustScaler - variáveis numéricas;\n",
        "  - Método CountEncoder - variáveis nominais;\n",
        "  - Método RandomUnderSampler - desbalanceamento da target.\n",
        "- Aplicação do modelo.\n",
        "- Análise de métrica de eficiência do modelo."
      ],
      "metadata": {
        "id": "qXylFlZig4Zo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Removendo coluna com valores de data/hora:\n",
        "base_amostra = base_amostra.drop(columns=['InvoiceDate', 'InvoiceHour', 'InvoiceHour_Intervalo']).copy()"
      ],
      "metadata": {
        "id": "HZ4sEl3Do3UE"
      },
      "execution_count": 144,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Separando variáveis categóricas e numéricas:\n",
        "cat_features = base_amostra.select_dtypes('object').drop(columns = 'Status').columns.to_list()\n",
        "num_features = base_amostra.select_dtypes(exclude = 'object').columns.to_list()\n",
        "\n",
        "cat_features, num_features"
      ],
      "metadata": {
        "id": "vBQtG9kYblCE",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e8dce0c9-b1f0-49d6-8392-d79d53365998"
      },
      "execution_count": 145,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(['InvoiceNo',\n",
              "  'StockCode',\n",
              "  'Description',\n",
              "  'CustomerID',\n",
              "  'Country',\n",
              "  'Continent'],\n",
              " ['Quantity', 'UnitPrice', 'TotalPrice'])"
            ]
          },
          "metadata": {},
          "execution_count": 145
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Divisão em dados de treino e de teste:\n",
        "X = base_amostra.drop(columns = ['Status'])\n",
        "y = base_amostra.Status.map({'Não Cancelado': 0,'Cancelado': 1})"
      ],
      "metadata": {
        "id": "VsDHIPwkf1pC"
      },
      "execution_count": 146,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# separa em treino e teste\n",
        "SEED = 546744165\n",
        "X_train, X_test, y_train, y_test = train_test_split(X,\n",
        "                                                     y,\n",
        "                                                     test_size = 0.3,\n",
        "                                                     random_state = SEED,\n",
        "                                                     stratify = y) #explicar"
      ],
      "metadata": {
        "id": "JugjXHdAaQFJ"
      },
      "execution_count": 147,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Distribuição da target antes do tratamento:\n",
        "pd.Series(y_test).value_counts().plot(kind = 'bar', color=colors)"
      ],
      "metadata": {
        "id": "q6crCnr1Rl48",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "outputId": "a9ee36ca-04c9-44f3-ab90-da77c3b94bea"
      },
      "execution_count": 148,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 148
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# from imblearn.under_sampling import RandomUnderSampler\n",
        "\n",
        "rus = RandomUnderSampler(random_state=SEED)\n",
        "X_res, y_res = rus.fit_resample(X_train, y_train)\n",
        "X_res_test, y_res_test = rus.fit_resample(X_test, y_test)"
      ],
      "metadata": {
        "id": "zo7-JjYHRYGM"
      },
      "execution_count": 149,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Distribuição da target após do tratamento:\n",
        "pd.Series(y_res).value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wfyipiVnYEPx",
        "outputId": "9214ad62-f6f8-41a4-a649-a46a69e9feb3"
      },
      "execution_count": 150,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    1592\n",
              "1    1592\n",
              "Name: Status, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 150
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.Series(y_res).value_counts().plot(kind = 'bar', color=colors)"
      ],
      "metadata": {
        "id": "6A422Xp-STdF",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "outputId": "61173896-e2e6-44eb-9021-5e7db717f438"
      },
      "execution_count": 151,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 151
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# from imblearn.over_sampling import SMOTE\n",
        "# from imblearn.over_sampling import SMOTENC\n",
        "\n",
        "# # smote = SMOTE()\n",
        "# categorical_features = [0, 1, 2, 5, 6, 8]\n",
        "# smotenc = SMOTENC(categorical_features=categorical_features)\n",
        "\n",
        "# X_train_resampled, y_train_resampled = smotenc.fit_resample(X_train, y_train)\n",
        "# X_test_resampled, y_test_resampled = smotenc.fit_resample(X_test, y_test)"
      ],
      "metadata": {
        "id": "vQ8ZM8xnZ5fa"
      },
      "execution_count": 152,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Distribuição da target após do tratamento:\n",
        "pd.Series(y_res).value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QyUETdtpaZY3",
        "outputId": "6578040f-9028-457f-fcde-b2c5e6f170f4"
      },
      "execution_count": 153,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    1592\n",
              "1    1592\n",
              "Name: Status, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 153
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# criar o transformador para aplicar as transformações:\n",
        "transformer = ColumnTransformer(\n",
        "     transformers=[\n",
        "          ('count_encoder', CountEncoder(), cat_features),\n",
        "          ('robust_scaler', RobustScaler(), num_features)],\n",
        "           remainder='passthrough')\n",
        "\n",
        "# criando o pipeline com o transformador e o modelo:\n",
        "pipeline = Pipeline([\n",
        "       ('transformer', transformer),\n",
        "       ('model', LogisticRegression())])\n",
        "\n",
        "# ajustar o pipeline aos dados de treino:\n",
        "pipeline.fit(X_res, y_res)"
      ],
      "metadata": {
        "id": "RNIAsvHMSZs9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 191
        },
        "outputId": "3b221ca1-bd3c-4429-a5e4-40477d030f8c"
      },
      "execution_count": 154,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('transformer',\n",
              "                 ColumnTransformer(remainder='passthrough',\n",
              "                                   transformers=[('count_encoder',\n",
              "                                                  CountEncoder(combine_min_nan_groups=True),\n",
              "                                                  ['InvoiceNo', 'StockCode',\n",
              "                                                   'Description', 'CustomerID',\n",
              "                                                   'Country', 'Continent']),\n",
              "                                                 ('robust_scaler',\n",
              "                                                  RobustScaler(),\n",
              "                                                  ['Quantity', 'UnitPrice',\n",
              "                                                   'TotalPrice'])])),\n",
              "                ('model', LogisticRegression())])"
            ],
            "text/html": [
              "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;transformer&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   transformers=[(&#x27;count_encoder&#x27;,\n",
              "                                                  CountEncoder(combine_min_nan_groups=True),\n",
              "                                                  [&#x27;InvoiceNo&#x27;, &#x27;StockCode&#x27;,\n",
              "                                                   &#x27;Description&#x27;, &#x27;CustomerID&#x27;,\n",
              "                                                   &#x27;Country&#x27;, &#x27;Continent&#x27;]),\n",
              "                                                 (&#x27;robust_scaler&#x27;,\n",
              "                                                  RobustScaler(),\n",
              "                                                  [&#x27;Quantity&#x27;, &#x27;UnitPrice&#x27;,\n",
              "                                                   &#x27;TotalPrice&#x27;])])),\n",
              "                (&#x27;model&#x27;, LogisticRegression())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-10\" type=\"checkbox\" ><label for=\"sk-estimator-id-10\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;transformer&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   transformers=[(&#x27;count_encoder&#x27;,\n",
              "                                                  CountEncoder(combine_min_nan_groups=True),\n",
              "                                                  [&#x27;InvoiceNo&#x27;, &#x27;StockCode&#x27;,\n",
              "                                                   &#x27;Description&#x27;, &#x27;CustomerID&#x27;,\n",
              "                                                   &#x27;Country&#x27;, &#x27;Continent&#x27;]),\n",
              "                                                 (&#x27;robust_scaler&#x27;,\n",
              "                                                  RobustScaler(),\n",
              "                                                  [&#x27;Quantity&#x27;, &#x27;UnitPrice&#x27;,\n",
              "                                                   &#x27;TotalPrice&#x27;])])),\n",
              "                (&#x27;model&#x27;, LogisticRegression())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" ><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">transformer: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                  transformers=[(&#x27;count_encoder&#x27;,\n",
              "                                 CountEncoder(combine_min_nan_groups=True),\n",
              "                                 [&#x27;InvoiceNo&#x27;, &#x27;StockCode&#x27;, &#x27;Description&#x27;,\n",
              "                                  &#x27;CustomerID&#x27;, &#x27;Country&#x27;, &#x27;Continent&#x27;]),\n",
              "                                (&#x27;robust_scaler&#x27;, RobustScaler(),\n",
              "                                 [&#x27;Quantity&#x27;, &#x27;UnitPrice&#x27;, &#x27;TotalPrice&#x27;])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">count_encoder</label><div class=\"sk-toggleable__content\"><pre>[&#x27;InvoiceNo&#x27;, &#x27;StockCode&#x27;, &#x27;Description&#x27;, &#x27;CustomerID&#x27;, &#x27;Country&#x27;, &#x27;Continent&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-13\" type=\"checkbox\" ><label for=\"sk-estimator-id-13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">CountEncoder</label><div class=\"sk-toggleable__content\"><pre>CountEncoder(combine_min_nan_groups=True)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-14\" type=\"checkbox\" ><label for=\"sk-estimator-id-14\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">robust_scaler</label><div class=\"sk-toggleable__content\"><pre>[&#x27;Quantity&#x27;, &#x27;UnitPrice&#x27;, &#x27;TotalPrice&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-15\" type=\"checkbox\" ><label for=\"sk-estimator-id-15\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RobustScaler</label><div class=\"sk-toggleable__content\"><pre>RobustScaler()</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-16\" type=\"checkbox\" ><label for=\"sk-estimator-id-16\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre>[]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-17\" type=\"checkbox\" ><label for=\"sk-estimator-id-17\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-18\" type=\"checkbox\" ><label for=\"sk-estimator-id-18\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression()</pre></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 154
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Prever os valores de saída para os dados de teste e treino\n",
        "y_pred_train = pipeline.predict(X_res)\n",
        "y_pred_test = pipeline.predict(X_res_test)\n",
        "\n",
        "# Probabilidade de saída para os dados de teste e treino\n",
        "y_pred_proba_train = pipeline.predict_proba(X_res)\n",
        "y_pred_proba_test = pipeline.predict_proba(X_res_test)"
      ],
      "metadata": {
        "id": "D7VIsMJETdC6"
      },
      "execution_count": 155,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Probabilidade de 0 e 1 para dados de treino\n",
        "proba_df = pd.DataFrame(y_pred_proba_train).round(5)\n",
        "proba_df['Real'] = y_res\n",
        "proba_df"
      ],
      "metadata": {
        "id": "NFM2rojZRgSq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "outputId": "048dc6ec-f766-41da-b77a-2ef0f19521e6"
      },
      "execution_count": 156,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "            0        1  Real\n",
              "0     0.99993  0.00007     0\n",
              "1     0.86464  0.13536     0\n",
              "2     0.98229  0.01771     0\n",
              "3     0.86227  0.13773     0\n",
              "4     0.99999  0.00001     0\n",
              "...       ...      ...   ...\n",
              "3179  0.00028  0.99972     1\n",
              "3180  0.00000  1.00000     1\n",
              "3181  0.00000  1.00000     1\n",
              "3182  0.00000  1.00000     1\n",
              "3183  0.00000  1.00000     1\n",
              "\n",
              "[3184 rows x 3 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-07906b6a-989e-477e-9678-b42a3d4ae99f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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",
              "      <th>1</th>\n",
              "      <th>Real</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.99993</td>\n",
              "      <td>0.00007</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.86464</td>\n",
              "      <td>0.13536</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.98229</td>\n",
              "      <td>0.01771</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.86227</td>\n",
              "      <td>0.13773</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.99999</td>\n",
              "      <td>0.00001</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3179</th>\n",
              "      <td>0.00028</td>\n",
              "      <td>0.99972</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3180</th>\n",
              "      <td>0.00000</td>\n",
              "      <td>1.00000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3181</th>\n",
              "      <td>0.00000</td>\n",
              "      <td>1.00000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3182</th>\n",
              "      <td>0.00000</td>\n",
              "      <td>1.00000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3183</th>\n",
              "      <td>0.00000</td>\n",
              "      <td>1.00000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3184 rows × 3 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-07906b6a-989e-477e-9678-b42a3d4ae99f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-312696a2-a101-4f50-921b-1b342a7a2cbb\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-312696a2-a101-4f50-921b-1b342a7a2cbb')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-312696a2-a101-4f50-921b-1b342a7a2cbb button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-07906b6a-989e-477e-9678-b42a3d4ae99f button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-07906b6a-989e-477e-9678-b42a3d4ae99f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 156
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.Series(y_pred_test).value_counts().plot(kind = 'bar', color=colors)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "id": "r0nIsk9Pwkqv",
        "outputId": "85792b6b-f8f7-4fc2-bac0-15adcbcc36f9"
      },
      "execution_count": 157,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 157
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### **Avaliando o modelo:**\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "qvHuExbiqdX7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Para fim didático, iremos utilizar as seguintes métricas para validação do modelo:\n",
        "\n",
        "- **`Matriz de Confusão:`** A matriz de confusão é uma representação tabular dos resultados de classificação de um modelo. Ela mostra a contagem de verdadeiros positivos, verdadeiros negativos, falsos positivos e falsos negativos, permitindo uma análise mais detalhada do desempenho do modelo.\n",
        "\n",
        "- **`Acurácia:`** A acurácia é uma métrica que mede a taxa de acertos do modelo, ou seja, a proporção de observações corretamente classificadas em relação ao total de observações. É uma medida geral do desempenho do modelo, indicando sua **capacidade de fazer previsões corretas**.\n",
        "\n",
        "- **`Precision (Precisão):`** A precisão é uma métrica que mede a proporção de observações positivas corretamente identificadas em relação ao total de observações classificadas como positivas. Ela é útil para avaliar a capacidade do modelo em evitar falsos positivos, ou seja, **classificar corretamente as instâncias positivas**.\n",
        "\n",
        "- **`Recall (Revocação):`** O recall é uma métrica que mede a proporção de observações positivas corretamente identificadas em relação ao total de observações positivas reais. Ela é útil para avaliar a capacidade do modelo em encontrar corretamente as instâncias positivas, **minimizando os falsos negativos**.\n",
        "\n",
        "- **`Curva ROC:`** A curva ROC representa a relação entre a taxa de verdadeiros positivos (TPR) e a taxa de falsos positivos (FPR) em diferentes pontos de corte (limiares de decisão) do modelo. O limiar de decisão é um valor ou critério que determina em qual classe o modelo irá classificar uma instância de entrada. Geralmente, o modelo calcula uma probabilidade de pertencer a uma determinada classe e compara essa probabilidade com o limiar de decisão para fazer a classificação final. Um bom modelo de classificação geralmente apresenta uma curva ROC **próxima ao canto superior esquerdo do gráfico, indicando uma alta taxa de verdadeiros positivos (sensibilidade) e uma baixa taxa de falsos positivos**. O ponto ideal seria aquele em que o modelo alcança uma taxa de verdadeiros positivos alta, minimizando simultaneamente a taxa de falsos positivos.\n",
        "\n",
        "- **`AUC-ROC:`** A AUC-ROC representa a probabilidade de o modelo classificar aleatoriamente um exemplo positivo mais alto do que um exemplo negativo. Quanto maior a AUC-ROC, melhor o desempenho do modelo em discriminar entre as classes. Esse valor representa a área sobre a curva ROC."
      ],
      "metadata": {
        "id": "8y2L0wbmW6vf"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "###### **Matriz de Confusão:**\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "-cFUt_lZXOCS"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Para interpretar a Matriz de Confusão, vamos considerar, a partir dos nosso dados:\n",
        "\n",
        "> - Valor Positivo: Cancelado (1)\n",
        "> - Valor Negativo: Não Cancelado (0)\n",
        "\n",
        "A Matriz de Confusão trás os valores previstos comparados aos valores reais da base, assim:\n",
        "\n",
        "> - Vedadeiro: Quando o modelo previu corretamente.\n",
        "> - Falso: Quando o modelo não previu corretamente.\n",
        "\n",
        "<img src = 'https://miro.medium.com/v2/resize:fit:1400/0*KUYgSbo0p1-5RRZW.png'/>\n",
        "\n",
        "Dessa forma, no quadrante de Verdadeiro Positivo (VP), temos a quantidade de previsões corretas de \"Cancelado\" (1), enquanto no quadrante de Verdadeiro Negativo (VN), temos a quantidade de previsões corretas de \"Não Cancelado\" (0).\n",
        "\n",
        "A mesma lógica se aplica aos quadrantes de Falso Positivo e Falso Negativo, representando as previsões incorretas."
      ],
      "metadata": {
        "id": "ULp1w1nHptA1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# matriz de confusão\n",
        "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
        "class_labels = [\"Não Cancelado\", \"Cancelado\"]\n",
        "ConfusionMatrixDisplay(confusion_matrix=confusion_matrix(y_res_test, y_pred_test), display_labels=class_labels).plot()\n",
        "plt.grid(False)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 458
        },
        "id": "fj1Nx6hbxPZ7",
        "outputId": "dfb3bc0a-08b1-404d-bc0d-1c618294f844"
      },
      "execution_count": 159,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Ao analisar a matriz de confusão acima, podemos observar que o modelo realizou 23 previsões incorretas de Falso Positivo, ou seja, classificou erroneamente 23 casos como \"Cancelado\" quando na verdade eram \"Não Cancelado\".\n",
        "\n",
        "Além disso, ocorreram 6 previsões incorretas de Falso Negativo, em que o modelo categorizou erroneamente 6 casos como \"Não Cancelado\" quando na verdade eram \"Cancelado\"."
      ],
      "metadata": {
        "id": "dCOhr4vruCSm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "###### **Acurácia, Precisão e Recall:**\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "RLBq_8XSzZ5V"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# cáculo precision, recall, f1 e acurácia\n",
        "from sklearn.metrics import classification_report\n",
        "\n",
        "report = classification_report(y_res_test, y_pred_test)\n",
        "print(report)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HoHEtGyNtaJD",
        "outputId": "7e1a4fbf-5684-4f47-abc3-73689ad8af89"
      },
      "execution_count": 160,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.97      0.99      0.98       683\n",
            "           1       0.99      0.97      0.98       683\n",
            "\n",
            "    accuracy                           0.98      1366\n",
            "   macro avg       0.98      0.98      0.98      1366\n",
            "weighted avg       0.98      0.98      0.98      1366\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###### **Curva ROC e AUC-ROC:**\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "Z0iWh_fu0knG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# cálculo da área da cuva ROC\n",
        "y_pred_proba_train = pipeline.predict_proba(X_res)[:, 1]  # probabilidades da classe positiva para o conjunto de treinamento\n",
        "y_pred_proba_test = pipeline.predict_proba(X_res_test)[:, 1]  # probabilidades da classe positiva para o conjunto de teste\n",
        "\n",
        "y_train_score = roc_auc_score(y_res, y_pred_proba_train)\n",
        "y_test_score = roc_auc_score(y_res_test, y_pred_proba_test)\n",
        "\n",
        "print(y_train_score, y_test_score)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YxAV4mbMxqgu",
        "outputId": "7808872b-ddf1-4889-9fc5-f30211d781c6"
      },
      "execution_count": 161,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9994089480063635 0.9985551642160908\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plotando curva ROC\n",
        "from sklearn.metrics import roc_curve\n",
        "\n",
        "fpr, tpr, thresholds = roc_curve(y_res_test, y_pred_proba_test)\n",
        "\n",
        "sns.set()\n",
        "sns.set_style(\"whitegrid\")\n",
        "\n",
        "plt.plot(fpr, tpr)\n",
        "plt.plot([0, 1], [0, 1], 'k--')  # linha diagonal para referência\n",
        "plt.xlabel('Taxa de Falsos Positivos')\n",
        "plt.ylabel('Taxa de Verdadeiros Positivos')\n",
        "plt.title('Curva ROC - Dados de Teste')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 480
        },
        "id": "YHlm2dSxzKBA",
        "outputId": "73d86088-4f29-4ff6-d4f8-c8f3970da874"
      },
      "execution_count": 162,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### ***Conclusões:***\n",
        "\n",
        "Com base nos resultados da métrica de classificação (precision, recall) e da acurácia apresentados acima, podemos concluir que o modelo possui um desempenho excelente na tarefa de classificação das observações em duas classes: 0 e 1.\n",
        "\n",
        "- A precisão média para ambas as classes é de 0.98, o que indica que o modelo tem uma alta taxa de acerto na classificação correta das observações positiva da base. **Ou seja, de tudo que o modelo classificou como \"Cancelado\", ele acertou 98%.**\n",
        "\n",
        "- O recall médio de 0.98 indica que o modelo é capaz de identificar corretamente a grande maioria das observações positivas em relação ao total de observações positivas presentes na base de dados. **Ou seja, o modelo consegue detectar cerca de 98% dos casos que realmente são \"Cancelados\".**\n",
        "\n",
        "- A acurácia geral do modelo é de 0.98, o que indica uma taxa de acerto elevada na classificação de todas as observações. Isso demonstra que **o modelo é capaz de realizar previsões corretas na maioria dos casos.**"
      ],
      "metadata": {
        "id": "yXQizBzcudxk"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Com base nos valores de AUC-ROC fornecidos, observamos que tanto no conjunto de treinamento quanto no conjunto de teste, o modelo apresenta um desempenho extremamente bom na tarefa de classificação. A AUC-ROC próxima de 1 indica que o modelo é capaz de fazer uma distinção clara entre as classes de cancelamento e não cancelamento de vendas, **com uma taxa muito baixa de falsos positivos em relação aos verdadeiros positivos**."
      ],
      "metadata": {
        "id": "yZT7vHqkQ1u3"
      }
    }
  ]
}