tabular_data_classifiers.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "tabular_data_classifiers.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPn8xylWV3elkk6azlbBIBs",
      "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/shllgtca/tabular_data_classifiers/blob/main/tabular_data_classifiers.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Setup**"
      ],
      "metadata": {
        "id": "J4cfFR621ozQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install dython > /dev/null\n",
        "!python -m pip uninstall -y matplotlib > /dev/null\n",
        "!pip install matplotlib==3.4.3 > /dev/null\n",
        "import os\n",
        "os.kill(os.getpid(), 9)"
      ],
      "metadata": {
        "id": "qlFdMDf4VwWK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from collections import Counter\n",
        "from sklearn.datasets import make_classification\n",
        "from imblearn.combine import SMOTETomek \n",
        "from dython.nominal import identify_nominal_columns\n",
        "from dython.nominal import associations\n",
        "from xgboost import XGBClassifier\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from imblearn.pipeline import Pipeline\n",
        "from sklearn.model_selection import train_test_split as tts\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "import pickle"
      ],
      "metadata": {
        "id": "dIh-dWA4_Fqr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://shllgtca: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@github.com/shllgtca/tabular_data_classifiers.git"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7_btYd3F1ucj",
        "outputId": "a56795f9-eae1-4eb3-f392-bd8be539b4b0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "fatal: destination path 'tabular_data_classifiers' already exists and is not an empty directory.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!ls \\tabular_data_classifiers"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MbkddKLK_n3B",
        "outputId": "9ead5660-4e05-47e0-ca11-e132aaf491d6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "O_G_Equipment_Data.xlsx  README.md\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "wd = '/content/tabular_data_classifiers/'\n",
        "db = pd.read_excel(wd+'O_G_Equipment_Data.xlsx')  \n",
        "db.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "amNjaRKo_Nwg",
        "outputId": "2fa7f5e6-e187-4e58-f978-feacdd70b29a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Cycle  Preset_1  Preset_2  Temperature    Pressure  VibrationX  VibrationY  \\\n",
              "0      1         3         6    44.235186   47.657254   46.441769   64.820327   \n",
              "1      2         2         4    60.807234   63.172076   62.005951   80.714431   \n",
              "2      3         2         1    79.027536   83.032190   82.642110   98.254386   \n",
              "3      4         2         3    79.716242  100.508634  122.362321  121.363429   \n",
              "4      5         2         5    39.989054   51.764833   42.514302   61.037910   \n",
              "\n",
              "   VibrationZ  Frequency   Fail  \n",
              "0   66.454520  44.483250  False  \n",
              "1   81.246405  60.228715  False  \n",
              "2   98.785196  80.993479  False  \n",
              "3  118.652538  80.315567  False  \n",
              "4   50.716469  64.245166  False  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6d5506dd-4b12-4ad5-9760-3095c71347bf\">\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>Cycle</th>\n",
              "      <th>Preset_1</th>\n",
              "      <th>Preset_2</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Pressure</th>\n",
              "      <th>VibrationX</th>\n",
              "      <th>VibrationY</th>\n",
              "      <th>VibrationZ</th>\n",
              "      <th>Frequency</th>\n",
              "      <th>Fail</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>44.235186</td>\n",
              "      <td>47.657254</td>\n",
              "      <td>46.441769</td>\n",
              "      <td>64.820327</td>\n",
              "      <td>66.454520</td>\n",
              "      <td>44.483250</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>60.807234</td>\n",
              "      <td>63.172076</td>\n",
              "      <td>62.005951</td>\n",
              "      <td>80.714431</td>\n",
              "      <td>81.246405</td>\n",
              "      <td>60.228715</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>79.027536</td>\n",
              "      <td>83.032190</td>\n",
              "      <td>82.642110</td>\n",
              "      <td>98.254386</td>\n",
              "      <td>98.785196</td>\n",
              "      <td>80.993479</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>79.716242</td>\n",
              "      <td>100.508634</td>\n",
              "      <td>122.362321</td>\n",
              "      <td>121.363429</td>\n",
              "      <td>118.652538</td>\n",
              "      <td>80.315567</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>39.989054</td>\n",
              "      <td>51.764833</td>\n",
              "      <td>42.514302</td>\n",
              "      <td>61.037910</td>\n",
              "      <td>50.716469</td>\n",
              "      <td>64.245166</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6d5506dd-4b12-4ad5-9760-3095c71347bf')\"\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",
              "  <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-6d5506dd-4b12-4ad5-9760-3095c71347bf 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-6d5506dd-4b12-4ad5-9760-3095c71347bf');\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": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "db.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WdhK07usAKzk",
        "outputId": "e89c49bc-cc0c-44f5-d1ed-b18eb3ec6015"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(800, 10)"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Problem comprehension**\n",
        "\n",
        "The main **goal** of this exercise is outline **characteristics and parameters** that **indicates** the equipment is on **the verge of failing**. "
      ],
      "metadata": {
        "id": "hOpY2d3N0UK2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Data exploration**\n"
      ],
      "metadata": {
        "id": "KBusBecR0T_x"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 1 – Calculate how many times the equipment has failed"
      ],
      "metadata": {
        "id": "SvMO4RoBBvfW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "db[db.Fail==True].shape[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p05uLjeKBuRL",
        "outputId": "b6cea063-c84d-41d2-c68c-d83e0a2c013c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "66"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "R: The equipment has failed: 66 times."
      ],
      "metadata": {
        "id": "evqjlqqPCf4B"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 2 – Categorize equipment failures by setups configurations (preset 1 and preset 2):"
      ],
      "metadata": {
        "id": "lpI6feCSBsIv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "db[db.Fail==True].pivot_table(index=['Preset_1'],\n",
        "                              values=['Fail'],\n",
        "                              aggfunc=len).sort_values(by='Fail',\n",
        "                                                       ascending=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "bSEzariWJzCJ",
        "outputId": "eca3bc38-1a6d-4c15-b946-edea03049003"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          Fail\n",
              "Preset_1      \n",
              "1           27\n",
              "2           21\n",
              "3           18"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-47acd74f-a547-427f-9c8c-fff021d1000d\">\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>Fail</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Preset_1</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>27</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>18</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-47acd74f-a547-427f-9c8c-fff021d1000d')\"\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",
              "  <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-47acd74f-a547-427f-9c8c-fff021d1000d 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-47acd74f-a547-427f-9c8c-fff021d1000d');\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": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "db[db.Fail==True].pivot_table(index=['Preset_2'],\n",
        "                              values=['Fail'],\n",
        "                              aggfunc=len).sort_values(by='Fail',\n",
        "                                                       ascending=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 331
        },
        "id": "oTUOtl2_B8HE",
        "outputId": "07687122-dd6b-433b-a612-7ba2a8364f15"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          Fail\n",
              "Preset_2      \n",
              "5           12\n",
              "1           11\n",
              "2            9\n",
              "6            9\n",
              "7            9\n",
              "8            7\n",
              "3            6\n",
              "4            3"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3106f8de-49db-48ec-8a93-ea8fca7f9ab4\">\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>Fail</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Preset_2</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3106f8de-49db-48ec-8a93-ea8fca7f9ab4')\"\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",
              "  <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-3106f8de-49db-48ec-8a93-ea8fca7f9ab4 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-3106f8de-49db-48ec-8a93-ea8fca7f9ab4');\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": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "R: The equipment has failled more for Preset_1 value 1, and Preset_2 value 12. Although, it's important to note that the combination of these two features enables other observations such as, for Preset_1 value 1 combined with Preset_2 and values 2 it was observed a greater failure frequency (5) than other combinations.\n"
      ],
      "metadata": {
        "id": "V_IODxQFGeY4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "db[db.Fail==True].pivot_table(index=['Preset_1','Preset_2'],\n",
        "                              values=['Fail'],\n",
        "                              aggfunc=len).sort_values(by='Fail',\n",
        "                                                       ascending=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 802
        },
        "id": "5Xvzy3RTRnyz",
        "outputId": "e0c24f28-d5cb-43bb-af27-092e67b60864"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                   Fail\n",
              "Preset_1 Preset_2      \n",
              "1        2            5\n",
              "         5            5\n",
              "         1            4\n",
              "3        5            4\n",
              "2        8            4\n",
              "1        7            4\n",
              "2        1            4\n",
              "3        1            3\n",
              "2        6            3\n",
              "         5            3\n",
              "3        6            3\n",
              "         7            3\n",
              "1        6            3\n",
              "2        2            2\n",
              "1        8            2\n",
              "2        7            2\n",
              "1        4            2\n",
              "3        2            2\n",
              "         3            2\n",
              "1        3            2\n",
              "2        3            2\n",
              "         4            1\n",
              "3        8            1"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8044abfa-b3a1-4c25-807e-f6e6545e8860\">\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></th>\n",
              "      <th>Fail</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Preset_1</th>\n",
              "      <th>Preset_2</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th rowspan=\"3\" valign=\"top\">1</th>\n",
              "      <th>2</th>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <th>5</th>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <th>8</th>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <th>7</th>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <th>1</th>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">2</th>\n",
              "      <th>6</th>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">3</th>\n",
              "      <th>6</th>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <th>6</th>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <th>8</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <th>7</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <th>4</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">3</th>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <th>3</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">2</th>\n",
              "      <th>3</th>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <th>8</th>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8044abfa-b3a1-4c25-807e-f6e6545e8860')\"\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",
              "  <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-8044abfa-b3a1-4c25-807e-f6e6545e8860 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-8044abfa-b3a1-4c25-807e-f6e6545e8860');\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": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 3 – Categorize equipment failures by their nature/root cause according to parameter readings (temperature, pressure, and others)\n",
        "\n",
        "R: To infer the failure nature relation with the equipment parameters it's possible "
      ],
      "metadata": {
        "id": "rZFLFlWlB2Bn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install matplotlib==3.1.3 > /dev/null"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KsZXCkTBneS5",
        "outputId": "c0f2ea10-f8a4-4ca8-842b-31b89be29661"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
            "dython 0.7.1.post4 requires matplotlib>=3.4.3, but you have matplotlib 3.1.3 which is incompatible.\u001b[0m\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "db[db.Fail==True].boxplot(column=['Temperature','Pressure','VibrationX','VibrationY','VibrationZ','Frequency'], grid=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "8xKCIaHJs6bv",
        "outputId": "3ae87030-931f-4bc5-cb15-b02940d7aa48"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<AxesSubplot:>"
            ]
          },
          "metadata": {},
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# categorical_features=identify_nominal_columns(db)\n",
        "categorical_features=['Cycle', 'Preset_1', 'Preset_2']\n",
        "fg,ax=plt.subplots(figsize=(8,8))\n",
        "res=associations(db, nominal_columns='auto', numerical_columns=None, \n",
        "             mark_columns=False, nom_nom_assoc='cramer', \n",
        "             num_num_assoc='spearman',\n",
        "             ax=ax, annot=True, fmt='.2f', cmap=None, \n",
        "             sv_color='silver', cbar=True, vmax=1.0, vmin=None, plot=True, \n",
        "             compute_only=False, clustering=False, title=None, filename=None)\n",
        "res['corr']"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 847
        },
        "id": "eiSIJ0lDVRMq",
        "outputId": "a6a0361c-204d-473b-f8ea-26ea91174f52"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x576 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                Cycle  Preset_1  Preset_2  Temperature  Pressure  VibrationX  \\\n",
              "Cycle        1.000000  0.050875  0.072316     0.120752  0.099957    0.119107   \n",
              "Preset_1     0.050875  1.000000  0.055422     0.006979 -0.003436    0.026697   \n",
              "Preset_2     0.072316  0.055422  1.000000     0.006820 -0.004336   -0.019043   \n",
              "Temperature  0.120752  0.006979  0.006820     1.000000  0.561720    0.434100   \n",
              "Pressure     0.099957 -0.003436 -0.004336     0.561720  1.000000    0.514878   \n",
              "VibrationX   0.119107  0.026697 -0.019043     0.434100  0.514878    1.000000   \n",
              "VibrationY   0.090342 -0.040457 -0.004177     0.469367  0.360031    0.312341   \n",
              "VibrationZ   0.076705 -0.021800 -0.002882     0.492154  0.321688    0.578653   \n",
              "Frequency    0.020609 -0.054174  0.020748     0.260984  0.373515    0.092807   \n",
              "Fail         0.145643 -0.046683 -0.016945     0.296204  0.319064    0.275350   \n",
              "\n",
              "             VibrationY  VibrationZ  Frequency      Fail  \n",
              "Cycle          0.090342    0.076705   0.020609  0.145643  \n",
              "Preset_1      -0.040457   -0.021800  -0.054174 -0.046683  \n",
              "Preset_2      -0.004177   -0.002882   0.020748 -0.016945  \n",
              "Temperature    0.469367    0.492154   0.260984  0.296204  \n",
              "Pressure       0.360031    0.321688   0.373515  0.319064  \n",
              "VibrationX     0.312341    0.578653   0.092807  0.275350  \n",
              "VibrationY     1.000000    0.571503   0.459061  0.377140  \n",
              "VibrationZ     0.571503    1.000000   0.319363  0.335157  \n",
              "Frequency      0.459061    0.319363   1.000000  0.313674  \n",
              "Fail           0.377140    0.335157   0.313674  1.000000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fbce3222-8609-48b2-9080-0decbbfa2705\">\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>Cycle</th>\n",
              "      <th>Preset_1</th>\n",
              "      <th>Preset_2</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Pressure</th>\n",
              "      <th>VibrationX</th>\n",
              "      <th>VibrationY</th>\n",
              "      <th>VibrationZ</th>\n",
              "      <th>Frequency</th>\n",
              "      <th>Fail</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Cycle</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.050875</td>\n",
              "      <td>0.072316</td>\n",
              "      <td>0.120752</td>\n",
              "      <td>0.099957</td>\n",
              "      <td>0.119107</td>\n",
              "      <td>0.090342</td>\n",
              "      <td>0.076705</td>\n",
              "      <td>0.020609</td>\n",
              "      <td>0.145643</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Preset_1</th>\n",
              "      <td>0.050875</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.055422</td>\n",
              "      <td>0.006979</td>\n",
              "      <td>-0.003436</td>\n",
              "      <td>0.026697</td>\n",
              "      <td>-0.040457</td>\n",
              "      <td>-0.021800</td>\n",
              "      <td>-0.054174</td>\n",
              "      <td>-0.046683</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Preset_2</th>\n",
              "      <td>0.072316</td>\n",
              "      <td>0.055422</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.006820</td>\n",
              "      <td>-0.004336</td>\n",
              "      <td>-0.019043</td>\n",
              "      <td>-0.004177</td>\n",
              "      <td>-0.002882</td>\n",
              "      <td>0.020748</td>\n",
              "      <td>-0.016945</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Temperature</th>\n",
              "      <td>0.120752</td>\n",
              "      <td>0.006979</td>\n",
              "      <td>0.006820</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.561720</td>\n",
              "      <td>0.434100</td>\n",
              "      <td>0.469367</td>\n",
              "      <td>0.492154</td>\n",
              "      <td>0.260984</td>\n",
              "      <td>0.296204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pressure</th>\n",
              "      <td>0.099957</td>\n",
              "      <td>-0.003436</td>\n",
              "      <td>-0.004336</td>\n",
              "      <td>0.561720</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.514878</td>\n",
              "      <td>0.360031</td>\n",
              "      <td>0.321688</td>\n",
              "      <td>0.373515</td>\n",
              "      <td>0.319064</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationX</th>\n",
              "      <td>0.119107</td>\n",
              "      <td>0.026697</td>\n",
              "      <td>-0.019043</td>\n",
              "      <td>0.434100</td>\n",
              "      <td>0.514878</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.312341</td>\n",
              "      <td>0.578653</td>\n",
              "      <td>0.092807</td>\n",
              "      <td>0.275350</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationY</th>\n",
              "      <td>0.090342</td>\n",
              "      <td>-0.040457</td>\n",
              "      <td>-0.004177</td>\n",
              "      <td>0.469367</td>\n",
              "      <td>0.360031</td>\n",
              "      <td>0.312341</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.571503</td>\n",
              "      <td>0.459061</td>\n",
              "      <td>0.377140</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationZ</th>\n",
              "      <td>0.076705</td>\n",
              "      <td>-0.021800</td>\n",
              "      <td>-0.002882</td>\n",
              "      <td>0.492154</td>\n",
              "      <td>0.321688</td>\n",
              "      <td>0.578653</td>\n",
              "      <td>0.571503</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.319363</td>\n",
              "      <td>0.335157</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Frequency</th>\n",
              "      <td>0.020609</td>\n",
              "      <td>-0.054174</td>\n",
              "      <td>0.020748</td>\n",
              "      <td>0.260984</td>\n",
              "      <td>0.373515</td>\n",
              "      <td>0.092807</td>\n",
              "      <td>0.459061</td>\n",
              "      <td>0.319363</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.313674</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Fail</th>\n",
              "      <td>0.145643</td>\n",
              "      <td>-0.046683</td>\n",
              "      <td>-0.016945</td>\n",
              "      <td>0.296204</td>\n",
              "      <td>0.319064</td>\n",
              "      <td>0.275350</td>\n",
              "      <td>0.377140</td>\n",
              "      <td>0.335157</td>\n",
              "      <td>0.313674</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fbce3222-8609-48b2-9080-0decbbfa2705')\"\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",
              "  <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-fbce3222-8609-48b2-9080-0decbbfa2705 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-fbce3222-8609-48b2-9080-0decbbfa2705');\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": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(8,7))\n",
        "sns.heatmap(db[db.Fail==True].iloc[:,3:-1].corr(),ax=ax)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "id": "MXiqVIn266NR",
        "outputId": "0f77fe72-dd90-4f82-e39c-c2d5935216ca"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<AxesSubplot:>"
            ]
          },
          "metadata": {},
          "execution_count": 13
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x504 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "db[db.Fail==True].iloc[:,3:-1].corr(method='spearman')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "xxSDJn5o-UOp",
        "outputId": "e705d510-149f-4cd7-b495-ad2b9c088bed"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "             Temperature  Pressure  VibrationX  VibrationY  VibrationZ  \\\n",
              "Temperature     1.000000  0.301952    0.160923    0.016762    0.009832   \n",
              "Pressure        0.301952  1.000000   -0.151320    0.544265   -0.291681   \n",
              "VibrationX      0.160923 -0.151320    1.000000   -0.380148    0.398142   \n",
              "VibrationY      0.016762  0.544265   -0.380148    1.000000   -0.077257   \n",
              "VibrationZ      0.009832 -0.291681    0.398142   -0.077257    1.000000   \n",
              "Frequency      -0.061017  0.164012   -0.480722    0.221292   -0.405657   \n",
              "\n",
              "             Frequency  \n",
              "Temperature  -0.061017  \n",
              "Pressure      0.164012  \n",
              "VibrationX   -0.480722  \n",
              "VibrationY    0.221292  \n",
              "VibrationZ   -0.405657  \n",
              "Frequency     1.000000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-111c651d-8521-4627-b592-ae1323582904\">\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>Temperature</th>\n",
              "      <th>Pressure</th>\n",
              "      <th>VibrationX</th>\n",
              "      <th>VibrationY</th>\n",
              "      <th>VibrationZ</th>\n",
              "      <th>Frequency</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Temperature</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.301952</td>\n",
              "      <td>0.160923</td>\n",
              "      <td>0.016762</td>\n",
              "      <td>0.009832</td>\n",
              "      <td>-0.061017</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pressure</th>\n",
              "      <td>0.301952</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.151320</td>\n",
              "      <td>0.544265</td>\n",
              "      <td>-0.291681</td>\n",
              "      <td>0.164012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationX</th>\n",
              "      <td>0.160923</td>\n",
              "      <td>-0.151320</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.380148</td>\n",
              "      <td>0.398142</td>\n",
              "      <td>-0.480722</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationY</th>\n",
              "      <td>0.016762</td>\n",
              "      <td>0.544265</td>\n",
              "      <td>-0.380148</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.077257</td>\n",
              "      <td>0.221292</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationZ</th>\n",
              "      <td>0.009832</td>\n",
              "      <td>-0.291681</td>\n",
              "      <td>0.398142</td>\n",
              "      <td>-0.077257</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.405657</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Frequency</th>\n",
              "      <td>-0.061017</td>\n",
              "      <td>0.164012</td>\n",
              "      <td>-0.480722</td>\n",
              "      <td>0.221292</td>\n",
              "      <td>-0.405657</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-111c651d-8521-4627-b592-ae1323582904')\"\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",
              "  <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-111c651d-8521-4627-b592-ae1323582904 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-111c651d-8521-4627-b592-ae1323582904');\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": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# from scipy.stats import zscore\n",
        "\n",
        "# z_scores = zscore(db.iloc[:,3:-1])\n",
        "\n",
        "# abs_z_scores = np.abs(z_scores)\n",
        "# filtered_entries = (abs_z_scores < 3).all(axis=1)\n",
        "# db = db[filtered_entries]"
      ],
      "metadata": {
        "id": "tc3p8OT97dsT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Model definition & explanation**\n"
      ],
      "metadata": {
        "id": "R7euRfcL0T09"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 4 - Create a model using the technique you think is most appropriate and measure its performance"
      ],
      "metadata": {
        "id": "hd2qiLhnZA-l"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "R: \n",
        "\n",
        "**Goal** : In order to understand the characteristics and parameters that indicates the equipment is on the verge of failing, by this meaning, to categorize the equipment failures by their nature/root cause according to parameter readings (temperature, pressure, and others) it's interesting to not only predict it's failure, but also, relate it to it's causes, by this meaning, relate to the feature that conduct the model to failure.\n",
        "\n",
        "\n",
        "**Preprocessing** : Due the dataset imbalanced, it was proceed an undersampling technique so called tomek links in addition with oversampling, synthetic data generation techique, smote, in order to achieve class balance.\n",
        "\n",
        "**Classification** : The selected algorithms for the model of this task were Random Forest and XGBoost classifiers. Firstly, because both are ensemble methods, which is a technique that creates multiple models and combine these models to produce improved results. This can benefit the model once there are variables wich behave similarly, such as vibrations measures, especially, on the x and z axis, as shown by the boxplot and correlation matrix of these values on the exploratory analysis. As the mais difference of XGBoost is to give more importance to optimizing the cost function of the model while Random Forest give more preference to hyperparameters to optimize the model, it's expected that the Random Forest performs more regularly regarding the analised metrics, than XGBoost.\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "mNaNXbDBvHIj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "db.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "Wa5H1LmC22Pn",
        "outputId": "5711b00e-0f38-42da-b3e1-e82d7be6f1e0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Cycle  Preset_1  Preset_2  Temperature    Pressure  VibrationX  VibrationY  \\\n",
              "0      1         3         6    44.235186   47.657254   46.441769   64.820327   \n",
              "1      2         2         4    60.807234   63.172076   62.005951   80.714431   \n",
              "2      3         2         1    79.027536   83.032190   82.642110   98.254386   \n",
              "3      4         2         3    79.716242  100.508634  122.362321  121.363429   \n",
              "4      5         2         5    39.989054   51.764833   42.514302   61.037910   \n",
              "\n",
              "   VibrationZ  Frequency   Fail  \n",
              "0   66.454520  44.483250  False  \n",
              "1   81.246405  60.228715  False  \n",
              "2   98.785196  80.993479  False  \n",
              "3  118.652538  80.315567  False  \n",
              "4   50.716469  64.245166  False  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e6fb3e35-86ae-4421-bec9-49de066d4eb5\">\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>Cycle</th>\n",
              "      <th>Preset_1</th>\n",
              "      <th>Preset_2</th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Pressure</th>\n",
              "      <th>VibrationX</th>\n",
              "      <th>VibrationY</th>\n",
              "      <th>VibrationZ</th>\n",
              "      <th>Frequency</th>\n",
              "      <th>Fail</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>44.235186</td>\n",
              "      <td>47.657254</td>\n",
              "      <td>46.441769</td>\n",
              "      <td>64.820327</td>\n",
              "      <td>66.454520</td>\n",
              "      <td>44.483250</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>60.807234</td>\n",
              "      <td>63.172076</td>\n",
              "      <td>62.005951</td>\n",
              "      <td>80.714431</td>\n",
              "      <td>81.246405</td>\n",
              "      <td>60.228715</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>79.027536</td>\n",
              "      <td>83.032190</td>\n",
              "      <td>82.642110</td>\n",
              "      <td>98.254386</td>\n",
              "      <td>98.785196</td>\n",
              "      <td>80.993479</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>79.716242</td>\n",
              "      <td>100.508634</td>\n",
              "      <td>122.362321</td>\n",
              "      <td>121.363429</td>\n",
              "      <td>118.652538</td>\n",
              "      <td>80.315567</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>39.989054</td>\n",
              "      <td>51.764833</td>\n",
              "      <td>42.514302</td>\n",
              "      <td>61.037910</td>\n",
              "      <td>50.716469</td>\n",
              "      <td>64.245166</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e6fb3e35-86ae-4421-bec9-49de066d4eb5')\"\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",
              "  <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-e6fb3e35-86ae-4421-bec9-49de066d4eb5 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-e6fb3e35-86ae-4421-bec9-49de066d4eb5');\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": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "db.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SL9f1FlV24ff",
        "outputId": "21e28be3-1e86-40b2-f5fa-0f33689f6e61"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(800, 10)"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pct_tst = 0.30\n",
        "X_train, X_test, y_train, y_test = tts(db.iloc[:,:-1], db.iloc[:,-1], test_size=pct_tst, random_state=0)\n",
        "print('Original train dataset shape %s' % Counter(y_train))\n",
        "grids={}\n",
        "y_preds={}\n",
        "smt = SMOTETomek()\n",
        "clf_XGB =  XGBClassifier(objective= 'binary:logistic',\n",
        "                          nthread=4,\n",
        "                          seed=42)\n",
        "clf_RF = RandomForestClassifier(random_state=0)\n",
        "clfs=[clf_RF,clf_XGB]\n",
        "\n",
        "params_RF=[{'classification_RF__max_depth': list(range(2,15))+[None]},\n",
        "          {'classification_RF__min_samples_split': list(range(2,10))}]\n",
        "\n",
        "params_XGB=[{'classification_XGB__max_depth': range(2, 10, 1)},\n",
        "          {'classification_XGB__n_estimators': range(10, 220, 40)},\n",
        "          {'classification_XGB__learning_rate': np.arange(0.005,0.5,0.005)}]            \n",
        "\n",
        "params_clfs=[params_RF,params_XGB]\n",
        "params = [{'sampling__random_state': [1]},\n",
        "           {'sampling__sampling_strategy':['minority','not minority',\n",
        "                                          'not minority','all','auto']+list(np.arange(0.2,1,0.1))}]\n",
        "\n",
        "for clf,params_clf in zip(clfs,params_clfs):          \n",
        "  clf_name = list(params_clf[0].keys())[0].split('__')[0]\n",
        "  print(clf_name)\n",
        "  model = Pipeline([\n",
        "          ('sampling', smt),\n",
        "          (clf_name,clf)\n",
        "      ])\n",
        "\n",
        "  grid = GridSearchCV(model, params_clf, \n",
        "                      scoring = 'f1',\n",
        "                      n_jobs = -1,\n",
        "                      cv = 5,\n",
        "                      verbose=True)\n",
        "\n",
        "  grid.fit(X_train, y_train)\n",
        "\n",
        "  y_pred = grid.best_estimator_[clf_name].predict(X_test)\n",
        "  grids[clf_name]=grid\n",
        "  y_preds[clf_name]=y_pred\n",
        "  print(classification_report(y_test, y_pred))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H4NsekzecZqQ",
        "outputId": "b4229370-792a-405d-e3c0-e7a4153001bd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original train dataset shape Counter({False: 518, True: 42})\n",
            "classification_RF\n",
            "Fitting 5 folds for each of 22 candidates, totalling 110 fits\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       False       0.97      0.97      0.97       216\n",
            "        True       0.72      0.75      0.73        24\n",
            "\n",
            "    accuracy                           0.95       240\n",
            "   macro avg       0.85      0.86      0.85       240\n",
            "weighted avg       0.95      0.95      0.95       240\n",
            "\n",
            "classification_XGB\n",
            "Fitting 5 folds for each of 113 candidates, totalling 565 fits\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       False       0.97      0.97      0.97       216\n",
            "        True       0.71      0.71      0.71        24\n",
            "\n",
            "    accuracy                           0.94       240\n",
            "   macro avg       0.84      0.84      0.84       240\n",
            "weighted avg       0.94      0.94      0.94       240\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "best_model = grids['classification_RF']"
      ],
      "metadata": {
        "id": "CRbnjxZOPhr-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "best_model.estimator['sampling'].sampling_strategy"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "t6jYfMR5PrJn",
        "outputId": "950d7592-1b48-4bcd-f5b1-6f4079420f82"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'auto'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "col='Feature importance'\n",
        "fs = pd.DataFrame(best_model.best_estimator_['classification_RF'].feature_importances_,\n",
        "                  index=db.iloc[:,:-1].columns.values,\n",
        "                  columns=[col]).sort_values(by=col,ascending=False)\n",
        "fs"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 331
        },
        "id": "HuQb6-0Gbg8D",
        "outputId": "d29aa0dd-76b4-4b7d-826e-2f386a8d56e1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "             Feature importance\n",
              "VibrationY             0.328969\n",
              "Pressure               0.174408\n",
              "Frequency              0.162591\n",
              "VibrationZ             0.113922\n",
              "VibrationX             0.068945\n",
              "Cycle                  0.065237\n",
              "Temperature            0.062568\n",
              "Preset_2               0.014710\n",
              "Preset_1               0.008649"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-892dea8f-1821-42ff-817c-f2eb824eb02c\">\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>Feature importance</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>VibrationY</th>\n",
              "      <td>0.328969</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pressure</th>\n",
              "      <td>0.174408</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Frequency</th>\n",
              "      <td>0.162591</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationZ</th>\n",
              "      <td>0.113922</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VibrationX</th>\n",
              "      <td>0.068945</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Cycle</th>\n",
              "      <td>0.065237</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Temperature</th>\n",
              "      <td>0.062568</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Preset_2</th>\n",
              "      <td>0.014710</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Preset_1</th>\n",
              "      <td>0.008649</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-892dea8f-1821-42ff-817c-f2eb824eb02c')\"\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",
              "  <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-892dea8f-1821-42ff-817c-f2eb824eb02c 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-892dea8f-1821-42ff-817c-f2eb824eb02c');\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": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_res, y_res = best_model.estimator['sampling'].fit_resample(X_train, y_train)\n",
        "print('Resampled train dataset shape %s' % Counter(y_res))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2RcOr6lX4l4i",
        "outputId": "c9c7abbe-f0e6-41a4-aeb6-a2e1787b06e8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Resampled train dataset shape Counter({False: 518, True: 518})\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "filename = 'best_model_tabular_data_classifiers.sav'\n",
        "pickle.dump(best_model, open(wd+filename, 'wb'))"
      ],
      "metadata": {
        "id": "Z8nnSYLzM2JF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tit='Failure Confusion Matrix'\n",
        "def conf_mat_plot(tst,pred,tit):\n",
        "  fg, ax = plt.subplots()\n",
        "  labels = tst.drop_duplicates().values\n",
        "  cm = confusion_matrix(tst,\n",
        "                        pred,\n",
        "                        labels=labels)    \n",
        "  cp = sns.color_palette(\"ch:s=.25,rot=-.25\", as_cmap=True)\n",
        "  f = sns.heatmap(cm, annot=True,fmt=\".0f\",cmap=cp,ax=ax)\n",
        "\n",
        "  dic={False:'No failure',True: 'Failure'}\n",
        "  labels = np.vectorize(dic.get)(labels)\n",
        "\n",
        "  ax.set_xticklabels(labels,ha='center', va='center_baseline')\n",
        "  ax.set_yticklabels(labels, va='center_baseline',rotation=0)\n",
        "  ax.set_xlabel('Predicted labels')\n",
        "  ax.set_ylabel('True labels')\n",
        "  fg.suptitle(tit)\n",
        "  return fg\n",
        "for clf_name,y_pred in y_preds.items():\n",
        "  fg = conf_mat_plot(y_test,y_pred,tit+' of '+clf_name)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 591
        },
        "id": "IumImf6h_FLV",
        "outputId": "e381eb67-79ed-4b04-e929-47d00d7ea3f4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Results evaluation**"
      ],
      "metadata": {
        "id": "bg6VjYaF0-Dr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "To evaluate the quality of the model acccuracy, it was calculated the confusion matrix of the results. This matrix allow us to verify the accuracy for each class,although the imbalanced problem has been addressed before by the undersampling preprocessing strategy, it's possible to check other metrics, such as precision, recall and f1-score on the classification report.\n",
        "\n",
        "It's important to note that due to machine and time restrictions, this exercise focused mainly on few classification models and resampling techiniques, although, there are a number of other techiniques and parameters that could be set for a better fine-tunning.\n",
        "\n",
        "The precision, recall and f1-score resuts around 80%, for both models were promissing results, showing that these data can be modeled despite the reduced quantity of samples and features. \n",
        "\n",
        "Regarding the feature importance for the best model, as expected, the vibration Y, pressure and frequency, responds for 65%, of the feature importance, eliminating highly corelated variables such as vibrations X and Z from the 3 top important features. \n",
        "\n",
        "Yet, other informations from businees owners could also help to predict the vessel failure, such as maintance frequency, for example. Other features could be add such as vessel operation type, allowing to analyze other vessel type data. Also the equipments inside the vessel could be specified by their types, and grouped the types of materials mainly used, or function, wich is a very important information when analyzing pressure and temperature effect. These grouping and feature add could enlarge the dataset providing more information."
      ],
      "metadata": {
        "id": "iBnQNkaMXASV"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Production pipeline proposal**"
      ],
      "metadata": {
        "id": "8e6JtgHFYLPs"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The approach of the model implementation depends a lot on the architeture of the enviroment in dispose, since the data aquisition until the user/agent interaction. \n",
        "\n",
        "So I will approach two different stragies for cloud and on premisse infrastructure. \n",
        "\n",
        "For the on premise architeture, such as Oracle database, the input data could be aquire through cx_Oracle module wich enables access to Oracle Database by a dsn connection, for example. So that input data can be handled in python. The model, an object that saved as .sav can be load and instanced by the python application allowing the data to be inferred according the model. So, there could be generated a report advising maintance status based on the model predictions. The interface with the user and/or an agent could be done by any orchestrator such as prefect. So that the prediction could be done by an interval (scheduled) or by the user action. \n",
        "\n",
        "For the cloud architeture, such as microsoft databricks, a datalake such as azure datalake can be used to store the data from equipments. The datalake can be mounted on the databricsk env to a lower cost resource allocation. And a notebook with the recorded model, could generate (scheduled) the predictions and export a report for the final user.\n",
        "\n",
        "For a mixed infrastructure, such as data on premise and cloud modeling etc, the data factory can be used to scheduling and loading the data to the datalake.And the model traine can be attached to the on premise infrastructure by a container,such as docker.\n",
        "\n",
        "These proposals could also be implemented on linux envs, although with different tools (for example, google colab, instead databricks, google drive insted of azure datalake, and airflow, instead prefect).\n",
        "\n",
        "If there's an issue due on the inference moment, due to the data dimensionality. So it can be treated in batches, or intervals, in order to demand less computation. There can also be implememnted some paralell or distributed programming in order to achieve greater performance. There can be used some spark, on higher level, or other lower level paralell/distributed pogramming such as mpi4py, or cython ().\n",
        "\n",
        "Also, according with the datascience lifecycle, its expected that these model to be deprecated or that adjustaments to be released, once it's a data specific model, in terms of accuracy, so, it must be clear that this pipeline msut be a Continuous Integration e Continuous Delivery flow.\n",
        "\n",
        "*https://cx-oracle.readthedocs.io/en/latest/index.html\n"
      ],
      "metadata": {
        "id": "GMDSilB0YSVL"
      }
    }
  ]
}