Car_Model_Detection.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/belanasaikiran/Team-6-Big-Data-Project/blob/main/Car_Model_Detection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PaAXM_g_gXEK"
      },
      "source": [
        "# Car Model Detection:\n",
        "\n",
        "### Team 6:"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3uEvKsqaRA77"
      },
      "source": [
        "## Todo:\n",
        "- [x] Import the DataSet.\n",
        "- [x] Preprocessing\n",
        "- [ ] Feature Selection\n",
        "- [ ] Model Selection\n",
        "- [ ] Training the Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sTXcmwR0REvG",
        "outputId": "ed9e37e1-b0c6-4f70-9b28-b7de81785c74"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/gdrive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "KTEuipBbc1rR"
      },
      "outputs": [],
      "source": [
        "!unzip gdrive/MyDrive/DataSet/DataSet.zip > /dev/null # output to null"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 58,
      "metadata": {
        "id": "QDPKza5ud7kv",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 497
        },
        "outputId": "2058cae3-d8c6-4d35-bbeb-1291a98bdb86"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=640x480>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 58
        }
      ],
      "source": [
        "import tensorflow as tf\n",
        "img = tf.keras.preprocessing.image.load_img('DataSet/images/cars_train/00010.jpg')\n",
        "img"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(img.width, img.height)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KnPpvSJ3mccY",
        "outputId": "56a85ba5-9830-48e5-9c28-6d653ec5aaab"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "640 480\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 52,
      "metadata": {
        "id": "cwPChQdPfwGQ"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "p7yXq6FUIvtn"
      },
      "outputs": [],
      "source": [
        "metadata = pd.read_csv(\"DataSet/data/meta.csv\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "vnySNtxYI8fW",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "b9051115-9af3-4ee7-dee6-4bf0282d2919"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Unnamed: 0                  class_name\n",
              "0           1  AM General Hummer SUV 2000\n",
              "1           2         Acura RL Sedan 2012\n",
              "2           3         Acura TL Sedan 2012\n",
              "3           4        Acura TL Type-S 2008\n",
              "4           5        Acura TSX Sedan 2012"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8c881a21-c141-496e-9c7b-1e9f76a6f5e1\" 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>Unnamed: 0</th>\n",
              "      <th>class_name</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>AM General Hummer SUV 2000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>Acura RL Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>Acura TL Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>Acura TL Type-S 2008</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>Acura TSX Sedan 2012</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8c881a21-c141-496e-9c7b-1e9f76a6f5e1')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-8c881a21-c141-496e-9c7b-1e9f76a6f5e1 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-8c881a21-c141-496e-9c7b-1e9f76a6f5e1');\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",
              "\n",
              "\n",
              "<div id=\"df-6554afdf-0dc0-4beb-a322-1e0bc320d575\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-6554afdf-0dc0-4beb-a322-1e0bc320d575')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-6554afdf-0dc0-4beb-a322-1e0bc320d575 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "metadata",
              "summary": "{\n  \"name\": \"metadata\",\n  \"rows\": 196,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 56,\n        \"min\": 1,\n        \"max\": 196,\n        \"num_unique_values\": 196,\n        \"samples\": [\n          140,\n          114,\n          17\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"class_name\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 196,\n        \"samples\": [\n          \"Hyundai Azera Sedan 2012\",\n          \"Ford F-150 Regular Cab 2007\",\n          \"Audi 100 Sedan 1994\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 18
        }
      ],
      "source": [
        "metadata.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "q3U_53IQJAvW",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e0d4deb7-0d31-41bd-fffd-1e61b0740fd8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 196 entries, 0 to 195\n",
            "Data columns (total 2 columns):\n",
            " #   Column      Non-Null Count  Dtype \n",
            "---  ------      --------------  ----- \n",
            " 0   Unnamed: 0  196 non-null    int64 \n",
            " 1   class_name  196 non-null    object\n",
            "dtypes: int64(1), object(1)\n",
            "memory usage: 3.2+ KB\n"
          ]
        }
      ],
      "source": [
        "metadata.info()"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "metadata.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WZhNW1YJnKOz",
        "outputId": "e1dc007d-3e32-490b-c3c4-5521a6acebe0"
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['Unnamed: 0', 'class_name'], dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Load the Train & Test Datasets"
      ],
      "metadata": {
        "id": "idyYVAlP9WcY"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "tsbx5WOwJH_E",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "a5c1299e-563d-4769-beaa-84aaf27bbf7c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Unnamed: 0   x1   y1    x2    y2  class      fname\n",
              "0           0   39  116   569   375     14  00001.jpg\n",
              "1           1   36  116   868   587      3  00002.jpg\n",
              "2           2   85  109   601   381     91  00003.jpg\n",
              "3           3  621  393  1484  1096    134  00004.jpg\n",
              "4           4   14   36   133    99    106  00005.jpg"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4b084b3b-20b5-4e69-8733-7b5138faf982\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>class</th>\n",
              "      <th>fname</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>39</td>\n",
              "      <td>116</td>\n",
              "      <td>569</td>\n",
              "      <td>375</td>\n",
              "      <td>14</td>\n",
              "      <td>00001.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>36</td>\n",
              "      <td>116</td>\n",
              "      <td>868</td>\n",
              "      <td>587</td>\n",
              "      <td>3</td>\n",
              "      <td>00002.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>85</td>\n",
              "      <td>109</td>\n",
              "      <td>601</td>\n",
              "      <td>381</td>\n",
              "      <td>91</td>\n",
              "      <td>00003.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>621</td>\n",
              "      <td>393</td>\n",
              "      <td>1484</td>\n",
              "      <td>1096</td>\n",
              "      <td>134</td>\n",
              "      <td>00004.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>14</td>\n",
              "      <td>36</td>\n",
              "      <td>133</td>\n",
              "      <td>99</td>\n",
              "      <td>106</td>\n",
              "      <td>00005.jpg</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4b084b3b-20b5-4e69-8733-7b5138faf982')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-4b084b3b-20b5-4e69-8733-7b5138faf982 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-4b084b3b-20b5-4e69-8733-7b5138faf982');\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",
              "\n",
              "\n",
              "<div id=\"df-6b7b2b9f-5947-4ffc-ae75-0cf27283fb3c\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-6b7b2b9f-5947-4ffc-ae75-0cf27283fb3c')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-6b7b2b9f-5947-4ffc-ae75-0cf27283fb3c button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "train_data",
              "summary": "{\n  \"name\": \"train_data\",\n  \"rows\": 12948,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2200,\n        \"min\": 0,\n        \"max\": 8143,\n        \"num_unique_values\": 8144,\n        \"samples\": [\n          7437,\n          483,\n          7598\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 81,\n        \"min\": 1,\n        \"max\": 1648,\n        \"num_unique_values\": 474,\n        \"samples\": [\n          471,\n          409,\n          21\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 106,\n        \"min\": 1,\n        \"max\": 1651,\n        \"num_unique_values\": 589,\n        \"samples\": [\n          481,\n          297,\n          541\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 409,\n        \"min\": 76,\n        \"max\": 5205,\n        \"num_unique_values\": 1643,\n        \"samples\": [\n          434,\n          356,\n          163\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 272,\n        \"min\": 41,\n        \"max\": 3389,\n        \"num_unique_values\": 1247,\n        \"samples\": [\n          763,\n          1282,\n          544\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"class\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 56,\n        \"min\": 1,\n        \"max\": 196,\n        \"num_unique_values\": 196,\n        \"samples\": [\n          7,\n          50,\n          73\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"fname\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 12948,\n        \"samples\": [\n          \"07278.jpg\",\n          \"05322.jpg\",\n          \"08421.jpg\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "train_data = pd.read_csv(\"DataSet/data/train.csv\")\n",
        "\n",
        "train_data.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "Hoz9kwrSJQQh",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "003ad32f-9545-4722-cc5c-f455e3264b1d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         Unnamed: 0            x1            y1            x2            y2  \\\n",
              "count  12948.000000  12948.000000  12948.000000  12948.000000  12948.000000   \n",
              "mean    3451.892339     64.701189    108.592833    638.598625    416.670683   \n",
              "std     2200.262018     81.774972    106.195279    409.613163    272.818630   \n",
              "min        0.000000      1.000000      1.000000     76.000000     41.000000   \n",
              "25%     1618.000000     19.000000     42.000000    394.000000    250.000000   \n",
              "50%     3236.500000     39.000000     80.000000    572.000000    361.000000   \n",
              "75%     4906.250000     78.000000    138.000000    742.250000    475.000000   \n",
              "max     8143.000000   1648.000000   1651.000000   5205.000000   3389.000000   \n",
              "\n",
              "              class  \n",
              "count  12948.000000  \n",
              "mean      99.008573  \n",
              "std       56.540538  \n",
              "min        1.000000  \n",
              "25%       50.000000  \n",
              "50%       99.000000  \n",
              "75%      148.000000  \n",
              "max      196.000000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1d67999e-80d7-4aad-b054-29b02078d5ae\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>12948.000000</td>\n",
              "      <td>12948.000000</td>\n",
              "      <td>12948.000000</td>\n",
              "      <td>12948.000000</td>\n",
              "      <td>12948.000000</td>\n",
              "      <td>12948.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>3451.892339</td>\n",
              "      <td>64.701189</td>\n",
              "      <td>108.592833</td>\n",
              "      <td>638.598625</td>\n",
              "      <td>416.670683</td>\n",
              "      <td>99.008573</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>2200.262018</td>\n",
              "      <td>81.774972</td>\n",
              "      <td>106.195279</td>\n",
              "      <td>409.613163</td>\n",
              "      <td>272.818630</td>\n",
              "      <td>56.540538</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>76.000000</td>\n",
              "      <td>41.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>1618.000000</td>\n",
              "      <td>19.000000</td>\n",
              "      <td>42.000000</td>\n",
              "      <td>394.000000</td>\n",
              "      <td>250.000000</td>\n",
              "      <td>50.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>3236.500000</td>\n",
              "      <td>39.000000</td>\n",
              "      <td>80.000000</td>\n",
              "      <td>572.000000</td>\n",
              "      <td>361.000000</td>\n",
              "      <td>99.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>4906.250000</td>\n",
              "      <td>78.000000</td>\n",
              "      <td>138.000000</td>\n",
              "      <td>742.250000</td>\n",
              "      <td>475.000000</td>\n",
              "      <td>148.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>8143.000000</td>\n",
              "      <td>1648.000000</td>\n",
              "      <td>1651.000000</td>\n",
              "      <td>5205.000000</td>\n",
              "      <td>3389.000000</td>\n",
              "      <td>196.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1d67999e-80d7-4aad-b054-29b02078d5ae')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-1d67999e-80d7-4aad-b054-29b02078d5ae 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-1d67999e-80d7-4aad-b054-29b02078d5ae');\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",
              "\n",
              "\n",
              "<div id=\"df-8ba32f29-33a9-4810-ac5a-0cd01b674729\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-8ba32f29-33a9-4810-ac5a-0cd01b674729')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-8ba32f29-33a9-4810-ac5a-0cd01b674729 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"train_data\",\n  \"rows\": 8,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4162.985412569492,\n        \"min\": 0.0,\n        \"max\": 12948.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          3451.89233858511,\n          3236.5,\n          12948.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4515.221827123711,\n        \"min\": 1.0,\n        \"max\": 12948.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          64.70118937287612,\n          39.0,\n          12948.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4504.2936541896615,\n        \"min\": 1.0,\n        \"max\": 12948.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          108.5928328699413,\n          80.0,\n          12948.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4492.990578997924,\n        \"min\": 76.0,\n        \"max\": 12948.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          638.598625270312,\n          572.0,\n          12948.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4449.888436545546,\n        \"min\": 41.0,\n        \"max\": 12948.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          416.6706827309237,\n          361.0,\n          12948.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"class\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4545.400466427845,\n        \"min\": 1.0,\n        \"max\": 12948.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          99.00857275254866,\n          99.0,\n          12948.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 10
        }
      ],
      "source": [
        "train_data.describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "mkpdb1ksJa19",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1d21363e-b339-4e62-a6be-27d795076145"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 12948 entries, 0 to 12947\n",
            "Data columns (total 7 columns):\n",
            " #   Column      Non-Null Count  Dtype \n",
            "---  ------      --------------  ----- \n",
            " 0   Unnamed: 0  12948 non-null  int64 \n",
            " 1   x1          12948 non-null  int64 \n",
            " 2   y1          12948 non-null  int64 \n",
            " 3   x2          12948 non-null  int64 \n",
            " 4   y2          12948 non-null  int64 \n",
            " 5   class       12948 non-null  int64 \n",
            " 6   fname       12948 non-null  object\n",
            "dtypes: int64(6), object(1)\n",
            "memory usage: 708.2+ KB\n"
          ]
        }
      ],
      "source": [
        "train_data.info()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "C8UZ2CW2JeKE",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d27627a5-dfe5-492e-9785-434c0eeac57a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(12948, 7)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ],
      "source": [
        "train_data.shape"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1E6I1V0n9edk",
        "outputId": "a67c7d71-0c96-454c-a68f-d02c2063a976"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['Unnamed: 0', 'x1', 'y1', 'x2', 'y2', 'class', 'fname'], dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_data = pd.read_csv(\"DataSet/data/test.csv\") #load test data\n",
        "test_data.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "tE312kT-9j61",
        "outputId": "35364fff-e178-421a-9064-63fc2a5a4ca9"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Unnamed: 0   x1   y1   x2   y2  class      fname\n",
              "0           0   30   52  246  147    181  00001.jpg\n",
              "1           1  100   19  576  203    103  00002.jpg\n",
              "2           2   51  105  968  659    145  00003.jpg\n",
              "3           3   67   84  581  407    187  00004.jpg\n",
              "4           4  140  151  593  339    185  00005.jpg"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-9e041e64-741d-4a51-a39d-1da11bbdfe2e\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>class</th>\n",
              "      <th>fname</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>30</td>\n",
              "      <td>52</td>\n",
              "      <td>246</td>\n",
              "      <td>147</td>\n",
              "      <td>181</td>\n",
              "      <td>00001.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>100</td>\n",
              "      <td>19</td>\n",
              "      <td>576</td>\n",
              "      <td>203</td>\n",
              "      <td>103</td>\n",
              "      <td>00002.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>51</td>\n",
              "      <td>105</td>\n",
              "      <td>968</td>\n",
              "      <td>659</td>\n",
              "      <td>145</td>\n",
              "      <td>00003.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>67</td>\n",
              "      <td>84</td>\n",
              "      <td>581</td>\n",
              "      <td>407</td>\n",
              "      <td>187</td>\n",
              "      <td>00004.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>140</td>\n",
              "      <td>151</td>\n",
              "      <td>593</td>\n",
              "      <td>339</td>\n",
              "      <td>185</td>\n",
              "      <td>00005.jpg</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9e041e64-741d-4a51-a39d-1da11bbdfe2e')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-9e041e64-741d-4a51-a39d-1da11bbdfe2e 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-9e041e64-741d-4a51-a39d-1da11bbdfe2e');\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",
              "\n",
              "\n",
              "<div id=\"df-37ce4517-0055-46de-9923-9a31ef3f7c79\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-37ce4517-0055-46de-9923-9a31ef3f7c79')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-37ce4517-0055-46de-9923-9a31ef3f7c79 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "test_data",
              "summary": "{\n  \"name\": \"test_data\",\n  \"rows\": 3237,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 934,\n        \"min\": 0,\n        \"max\": 3236,\n        \"num_unique_values\": 3237,\n        \"samples\": [\n          844,\n          2407,\n          1427\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 84,\n        \"min\": 1,\n        \"max\": 872,\n        \"num_unique_values\": 325,\n        \"samples\": [\n          221,\n          35,\n          179\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 106,\n        \"min\": 1,\n        \"max\": 1311,\n        \"num_unique_values\": 420,\n        \"samples\": [\n          113,\n          158,\n          252\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 417,\n        \"min\": 72,\n        \"max\": 7224,\n        \"num_unique_values\": 1034,\n        \"samples\": [\n          300,\n          834,\n          1133\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 279,\n        \"min\": 49,\n        \"max\": 3835,\n        \"num_unique_values\": 838,\n        \"samples\": [\n          238,\n          1061,\n          328\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"class\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 56,\n        \"min\": 1,\n        \"max\": 196,\n        \"num_unique_values\": 196,\n        \"samples\": [\n          153,\n          179,\n          154\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"fname\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3237,\n        \"samples\": [\n          \"00845.jpg\",\n          \"02408.jpg\",\n          \"01428.jpg\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Split Data into Training and Validation Set"
      ],
      "metadata": {
        "id": "f3nS3d579732"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "import tensorflow as tf"
      ],
      "metadata": {
        "id": "ttMUk5tV96UY"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train = train_data.drop(columns=[\"class\"]) #Features\n",
        "y_train = train_data[\"class\"] # Target variable which in our case is class"
      ],
      "metadata": {
        "id": "R3lCT67BKFSX"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.2, random_state=42) #training set 80% and validation set 20%"
      ],
      "metadata": {
        "id": "LxcvQ657KNbC"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BGGd5LUgKg1y",
        "outputId": "ed09750d-abe8-429d-c1d0-c57d53162a2c"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(10358, 6)"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_val.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e4x5e7HGKiTm",
        "outputId": "5323552d-6915-4507-9e2f-a0f3bf3d0930"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2590, 6)"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_val.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uGBCuQ6LKjuJ",
        "outputId": "f4c12a10-9630-4230-cd89-f4ba49b4eda4"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2590,)"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_train.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "NPug46aqKlc5",
        "outputId": "ee8cc3e8-1603-4e17-87ae-8d534c421884"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "      Unnamed: 0   x1   y1    x2   y2      fname\n",
              "6821        6821   22  186   975  652  06822.jpg\n",
              "2498        2498   49  163   733  438  02499.jpg\n",
              "1851        1851   27   21   626  433  01852.jpg\n",
              "8128        8128  529  463  1306  808  08129.jpg\n",
              "4533        4533   27   73   555  330  04534.jpg"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5be8fe9d-4633-4f4a-b5b6-2b206049d76c\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>fname</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>6821</th>\n",
              "      <td>6821</td>\n",
              "      <td>22</td>\n",
              "      <td>186</td>\n",
              "      <td>975</td>\n",
              "      <td>652</td>\n",
              "      <td>06822.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2498</th>\n",
              "      <td>2498</td>\n",
              "      <td>49</td>\n",
              "      <td>163</td>\n",
              "      <td>733</td>\n",
              "      <td>438</td>\n",
              "      <td>02499.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1851</th>\n",
              "      <td>1851</td>\n",
              "      <td>27</td>\n",
              "      <td>21</td>\n",
              "      <td>626</td>\n",
              "      <td>433</td>\n",
              "      <td>01852.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8128</th>\n",
              "      <td>8128</td>\n",
              "      <td>529</td>\n",
              "      <td>463</td>\n",
              "      <td>1306</td>\n",
              "      <td>808</td>\n",
              "      <td>08129.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4533</th>\n",
              "      <td>4533</td>\n",
              "      <td>27</td>\n",
              "      <td>73</td>\n",
              "      <td>555</td>\n",
              "      <td>330</td>\n",
              "      <td>04534.jpg</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5be8fe9d-4633-4f4a-b5b6-2b206049d76c')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-5be8fe9d-4633-4f4a-b5b6-2b206049d76c 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-5be8fe9d-4633-4f4a-b5b6-2b206049d76c');\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",
              "\n",
              "\n",
              "<div id=\"df-32103e62-ce20-4ebf-b550-0fa95c81b311\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-32103e62-ce20-4ebf-b550-0fa95c81b311')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-32103e62-ce20-4ebf-b550-0fa95c81b311 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "X_train",
              "summary": "{\n  \"name\": \"X_train\",\n  \"rows\": 10358,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2199,\n        \"min\": 0,\n        \"max\": 8143,\n        \"num_unique_values\": 7303,\n        \"samples\": [\n          1958,\n          2544,\n          1514\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 82,\n        \"min\": 1,\n        \"max\": 1648,\n        \"num_unique_values\": 451,\n        \"samples\": [\n          367,\n          257,\n          270\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 105,\n        \"min\": 1,\n        \"max\": 1646,\n        \"num_unique_values\": 559,\n        \"samples\": [\n          101,\n          1187,\n          106\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 410,\n        \"min\": 76,\n        \"max\": 5205,\n        \"num_unique_values\": 1534,\n        \"samples\": [\n          1530,\n          1582,\n          334\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 273,\n        \"min\": 41,\n        \"max\": 3389,\n        \"num_unique_values\": 1177,\n        \"samples\": [\n          385,\n          1039,\n          167\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"fname\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 10358,\n        \"samples\": [\n          \"05807.jpg\",\n          \"00375.jpg\",\n          \"09288.jpg\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_train_array = X_train.to_numpy()"
      ],
      "metadata": {
        "id": "ff89bCTeKnOe"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train_reshaped = X_train_array.reshape(10358, 28*28)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 141
        },
        "id": "38QBXrzjKtRa",
        "outputId": "119dbc59-271a-4f91-cbab-b01ab528d71c"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "cannot reshape array of size 62148 into shape (10358,784)",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-32-359af6d27b25>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mX_train_reshaped\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mX_train_array\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m10358\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m28\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0;36m28\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m: cannot reshape array of size 62148 into shape (10358,784)"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Prepare the Data for Machine Learning Algorithms"
      ],
      "metadata": {
        "id": "11AK2oWvLBVq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Data Cleaning"
      ],
      "metadata": {
        "id": "FRep6k4iLDmo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sample_incomplete_rows = train_data[train_data.isnull().any(axis=1)].head() #maybe we don't have any missing data\n",
        "pd.set_option('display.max_columns', None)\n",
        "sample_incomplete_rows"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "vhRUz58VKzff",
        "outputId": "3fe72e76-7146-4b1a-de7d-25f4234fbc5f"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Empty DataFrame\n",
              "Columns: [Unnamed: 0, x1, y1, x2, y2, class, fname]\n",
              "Index: []"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-98529d71-cb2a-4569-868d-e8dbcf4fcc67\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>class</th>\n",
              "      <th>fname</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-98529d71-cb2a-4569-868d-e8dbcf4fcc67')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-98529d71-cb2a-4569-868d-e8dbcf4fcc67 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-98529d71-cb2a-4569-868d-e8dbcf4fcc67');\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",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "sample_incomplete_rows",
              "repr_error": "Out of range float values are not JSON compliant: nan"
            }
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Calculate the width and height of bounding boxes"
      ],
      "metadata": {
        "id": "-i3u-NQFe9LX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate width of bounding boxes\n",
        "train_data['bbox_width'] = train_data['x2'] - train_data['x1']\n",
        "\n",
        "# Calculate height of bounding boxes\n",
        "train_data['bbox_height'] = train_data['y2'] - train_data['y1']\n",
        "\n",
        "# Print the first few rows to verify the new features\n",
        "print(train_data[['bbox_width', 'bbox_height']].head())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4jBUnsS-LF3A",
        "outputId": "af484c02-d935-4c73-cdde-2d0c2e0119c3"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "   bbox_width  bbox_height\n",
            "0         530          259\n",
            "1         832          471\n",
            "2         516          272\n",
            "3         863          703\n",
            "4         119           63\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "JHUeIZRNfDFA",
        "outputId": "8431f1ac-78cb-4a48-9473-57271a6d1e67"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Unnamed: 0   x1   y1    x2    y2  class      fname  bbox_width  bbox_height\n",
              "0           0   39  116   569   375     14  00001.jpg         530          259\n",
              "1           1   36  116   868   587      3  00002.jpg         832          471\n",
              "2           2   85  109   601   381     91  00003.jpg         516          272\n",
              "3           3  621  393  1484  1096    134  00004.jpg         863          703\n",
              "4           4   14   36   133    99    106  00005.jpg         119           63"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-10255566-0f1f-475d-b63f-0dcba1bd1c9a\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>class</th>\n",
              "      <th>fname</th>\n",
              "      <th>bbox_width</th>\n",
              "      <th>bbox_height</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>39</td>\n",
              "      <td>116</td>\n",
              "      <td>569</td>\n",
              "      <td>375</td>\n",
              "      <td>14</td>\n",
              "      <td>00001.jpg</td>\n",
              "      <td>530</td>\n",
              "      <td>259</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>36</td>\n",
              "      <td>116</td>\n",
              "      <td>868</td>\n",
              "      <td>587</td>\n",
              "      <td>3</td>\n",
              "      <td>00002.jpg</td>\n",
              "      <td>832</td>\n",
              "      <td>471</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>85</td>\n",
              "      <td>109</td>\n",
              "      <td>601</td>\n",
              "      <td>381</td>\n",
              "      <td>91</td>\n",
              "      <td>00003.jpg</td>\n",
              "      <td>516</td>\n",
              "      <td>272</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>621</td>\n",
              "      <td>393</td>\n",
              "      <td>1484</td>\n",
              "      <td>1096</td>\n",
              "      <td>134</td>\n",
              "      <td>00004.jpg</td>\n",
              "      <td>863</td>\n",
              "      <td>703</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>14</td>\n",
              "      <td>36</td>\n",
              "      <td>133</td>\n",
              "      <td>99</td>\n",
              "      <td>106</td>\n",
              "      <td>00005.jpg</td>\n",
              "      <td>119</td>\n",
              "      <td>63</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-10255566-0f1f-475d-b63f-0dcba1bd1c9a')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-10255566-0f1f-475d-b63f-0dcba1bd1c9a 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-10255566-0f1f-475d-b63f-0dcba1bd1c9a');\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",
              "\n",
              "\n",
              "<div id=\"df-219a1451-77a0-4584-a0f8-cf4e660e58d7\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-219a1451-77a0-4584-a0f8-cf4e660e58d7')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-219a1451-77a0-4584-a0f8-cf4e660e58d7 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "train_data",
              "summary": "{\n  \"name\": \"train_data\",\n  \"rows\": 12948,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2200,\n        \"min\": 0,\n        \"max\": 8143,\n        \"num_unique_values\": 8144,\n        \"samples\": [\n          7437,\n          483,\n          7598\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 81,\n        \"min\": 1,\n        \"max\": 1648,\n        \"num_unique_values\": 474,\n        \"samples\": [\n          471,\n          409,\n          21\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 106,\n        \"min\": 1,\n        \"max\": 1651,\n        \"num_unique_values\": 589,\n        \"samples\": [\n          481,\n          297,\n          541\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 409,\n        \"min\": 76,\n        \"max\": 5205,\n        \"num_unique_values\": 1643,\n        \"samples\": [\n          434,\n          356,\n          163\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 272,\n        \"min\": 41,\n        \"max\": 3389,\n        \"num_unique_values\": 1247,\n        \"samples\": [\n          763,\n          1282,\n          544\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"class\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 56,\n        \"min\": 1,\n        \"max\": 196,\n        \"num_unique_values\": 196,\n        \"samples\": [\n          7,\n          50,\n          73\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"fname\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 12948,\n        \"samples\": [\n          \"07278.jpg\",\n          \"05322.jpg\",\n          \"08421.jpg\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"bbox_width\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 374,\n        \"min\": 71,\n        \"max\": 4545,\n        \"num_unique_values\": 1555,\n        \"samples\": [\n          901,\n          596,\n          521\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"bbox_height\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 212,\n        \"min\": 23,\n        \"max\": 2895,\n        \"num_unique_values\": 1017,\n        \"samples\": [\n          811,\n          319,\n          705\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Calculate summary statistics for numerical features to determine should we apply normalization and Standardization"
      ],
      "metadata": {
        "id": "NbDGo7GFfIV2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate summary statistics for numerical features\n",
        "summary_statistics = train_data.describe()\n",
        "\n",
        "# Print summary statistics to check the scale of features\n",
        "print(summary_statistics)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0pISfLJDfGlo",
        "outputId": "99207da6-53ff-4fad-cdd9-97ba0a9cca59"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "         Unnamed: 0            x1            y1            x2            y2  \\\n",
            "count  12948.000000  12948.000000  12948.000000  12948.000000  12948.000000   \n",
            "mean    3451.892339     64.701189    108.592833    638.598625    416.670683   \n",
            "std     2200.262018     81.774972    106.195279    409.613163    272.818630   \n",
            "min        0.000000      1.000000      1.000000     76.000000     41.000000   \n",
            "25%     1618.000000     19.000000     42.000000    394.000000    250.000000   \n",
            "50%     3236.500000     39.000000     80.000000    572.000000    361.000000   \n",
            "75%     4906.250000     78.000000    138.000000    742.250000    475.000000   \n",
            "max     8143.000000   1648.000000   1651.000000   5205.000000   3389.000000   \n",
            "\n",
            "              class    bbox_width   bbox_height  \n",
            "count  12948.000000  12948.000000  12948.000000  \n",
            "mean      99.008573    573.897436    308.077850  \n",
            "std       56.540538    374.261681    212.308214  \n",
            "min        1.000000     71.000000     23.000000  \n",
            "25%       50.000000    343.000000    174.000000  \n",
            "50%       99.000000    508.000000    261.000000  \n",
            "75%      148.000000    656.000000    369.250000  \n",
            "max      196.000000   4545.000000   2895.000000  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Normalization and Standardization"
      ],
      "metadata": {
        "id": "d8F0Ps-Bi7QK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import MinMaxScaler\n",
        "\n",
        "# Initialize the MinMaxScaler\n",
        "scaler = MinMaxScaler()\n",
        "\n",
        "# Define the numerical features to be normalized\n",
        "numerical_features = ['x1', 'y1', 'x2', 'y2', 'class', 'bbox_width', 'bbox_height']\n",
        "\n",
        "# Apply normalization to the numerical features\n",
        "train_data_normalized = train_data.copy()  # Create a copy of the original DataFrame\n",
        "train_data_normalized[numerical_features] = scaler.fit_transform(train_data_normalized[numerical_features])\n",
        "\n",
        "train_data_normalized.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "yfjsdpSvi4-F",
        "outputId": "08da0d7a-b226-4fc0-b3d7-78551c5168ae"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Unnamed: 0        x1        y1        x2        y2     class      fname  \\\n",
              "0           0  0.023072  0.069697  0.096120  0.099761  0.066667  00001.jpg   \n",
              "1           1  0.021251  0.069697  0.154416  0.163082  0.010256  00002.jpg   \n",
              "2           2  0.051002  0.065455  0.102359  0.101553  0.461538  00003.jpg   \n",
              "3           3  0.376442  0.237576  0.274517  0.315114  0.682051  00004.jpg   \n",
              "4           4  0.007893  0.021212  0.011113  0.017324  0.538462  00005.jpg   \n",
              "\n",
              "   bbox_width  bbox_height  \n",
              "0    0.102593     0.082173  \n",
              "1    0.170094     0.155989  \n",
              "2    0.099464     0.086699  \n",
              "3    0.177023     0.236769  \n",
              "4    0.010729     0.013928  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3c57de06-fdbe-4893-8288-095c7f9bb67b\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>class</th>\n",
              "      <th>fname</th>\n",
              "      <th>bbox_width</th>\n",
              "      <th>bbox_height</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0.023072</td>\n",
              "      <td>0.069697</td>\n",
              "      <td>0.096120</td>\n",
              "      <td>0.099761</td>\n",
              "      <td>0.066667</td>\n",
              "      <td>00001.jpg</td>\n",
              "      <td>0.102593</td>\n",
              "      <td>0.082173</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>0.021251</td>\n",
              "      <td>0.069697</td>\n",
              "      <td>0.154416</td>\n",
              "      <td>0.163082</td>\n",
              "      <td>0.010256</td>\n",
              "      <td>00002.jpg</td>\n",
              "      <td>0.170094</td>\n",
              "      <td>0.155989</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>0.051002</td>\n",
              "      <td>0.065455</td>\n",
              "      <td>0.102359</td>\n",
              "      <td>0.101553</td>\n",
              "      <td>0.461538</td>\n",
              "      <td>00003.jpg</td>\n",
              "      <td>0.099464</td>\n",
              "      <td>0.086699</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>0.376442</td>\n",
              "      <td>0.237576</td>\n",
              "      <td>0.274517</td>\n",
              "      <td>0.315114</td>\n",
              "      <td>0.682051</td>\n",
              "      <td>00004.jpg</td>\n",
              "      <td>0.177023</td>\n",
              "      <td>0.236769</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>0.007893</td>\n",
              "      <td>0.021212</td>\n",
              "      <td>0.011113</td>\n",
              "      <td>0.017324</td>\n",
              "      <td>0.538462</td>\n",
              "      <td>00005.jpg</td>\n",
              "      <td>0.010729</td>\n",
              "      <td>0.013928</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3c57de06-fdbe-4893-8288-095c7f9bb67b')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-3c57de06-fdbe-4893-8288-095c7f9bb67b 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-3c57de06-fdbe-4893-8288-095c7f9bb67b');\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",
              "\n",
              "\n",
              "<div id=\"df-48e66334-e058-4fb1-a831-6aff2ef60493\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-48e66334-e058-4fb1-a831-6aff2ef60493')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-48e66334-e058-4fb1-a831-6aff2ef60493 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "train_data_normalized",
              "summary": "{\n  \"name\": \"train_data_normalized\",\n  \"rows\": 12948,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2200,\n        \"min\": 0,\n        \"max\": 8143,\n        \"num_unique_values\": 8144,\n        \"samples\": [\n          7437,\n          483,\n          7598\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.04965086311537254,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 474,\n        \"samples\": [\n          0.28536733454766244,\n          0.24772313296903462,\n          0.012143290831815423\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.06436077535566918,\n        \"min\": 0.0,\n        \"max\": 0.9999999999999999,\n        \"num_unique_values\": 589,\n        \"samples\": [\n          0.2909090909090909,\n          0.17939393939393938,\n          0.32727272727272727\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.07986218818468498,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 1643,\n        \"samples\": [\n          0.06979918112692532,\n          0.0545915383115617,\n          0.016962370832520956\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.08148704602608837,\n        \"min\": 0.0,\n        \"max\": 0.9999999999999999,\n        \"num_unique_values\": 1247,\n        \"samples\": [\n          0.21565113500597372,\n          0.3706690561529271,\n          0.1502389486260454\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"class\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.2899514786125128,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 196,\n        \"samples\": [\n          0.030769230769230767,\n          0.2512820512820513,\n          0.36923076923076925\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"fname\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 12948,\n        \"samples\": [\n          \"07278.jpg\",\n          \"05322.jpg\",\n          \"08421.jpg\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"bbox_width\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.08365258840210986,\n        \"min\": 0.0,\n        \"max\": 1.0000000000000002,\n        \"num_unique_values\": 1555,\n        \"samples\": [\n          0.18551631649530623,\n          0.11734465802413949,\n          0.10058113544926241\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"bbox_height\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.0739234729826433,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 1017,\n        \"samples\": [\n          0.2743732590529248,\n          0.10306406685236769,\n          0.23746518105849584\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "# Initialize the StandardScaler\n",
        "scaler = StandardScaler()\n",
        "\n",
        "# Apply standardization to the numerical features\n",
        "train_data_standardized = train_data.copy()  # Create a copy of the original DataFrame\n",
        "train_data_standardized[numerical_features] = scaler.fit_transform(train_data_standardized[numerical_features])\n",
        "\n",
        "train_data_standardized.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "4tQj7IVji-_i",
        "outputId": "792eb3cc-e63c-433d-cae3-a58cf3a30de9"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Unnamed: 0        x1        y1        x2        y2     class      fname  \\\n",
              "0           0 -0.314304  0.069753 -0.169920 -0.152747 -1.503556  00001.jpg   \n",
              "1           1 -0.350991  0.069753  0.560066  0.624356 -1.698114  00002.jpg   \n",
              "2           2  0.248237  0.003834 -0.091794 -0.130754 -0.141648  00003.jpg   \n",
              "3           3  6.803063  2.678256  2.063982  2.490136  0.618897  00004.jpg   \n",
              "4           4 -0.620033 -0.683605 -1.234380 -1.164447  0.123658  00005.jpg   \n",
              "\n",
              "   bbox_width  bbox_height  \n",
              "0   -0.117295    -0.231172  \n",
              "1    0.689658     0.767415  \n",
              "2   -0.154704    -0.169938  \n",
              "3    0.772491     1.860208  \n",
              "4   -1.215500    -1.154394  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5cbbdd73-8709-41ff-8419-b3893033986f\" 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>Unnamed: 0</th>\n",
              "      <th>x1</th>\n",
              "      <th>y1</th>\n",
              "      <th>x2</th>\n",
              "      <th>y2</th>\n",
              "      <th>class</th>\n",
              "      <th>fname</th>\n",
              "      <th>bbox_width</th>\n",
              "      <th>bbox_height</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>-0.314304</td>\n",
              "      <td>0.069753</td>\n",
              "      <td>-0.169920</td>\n",
              "      <td>-0.152747</td>\n",
              "      <td>-1.503556</td>\n",
              "      <td>00001.jpg</td>\n",
              "      <td>-0.117295</td>\n",
              "      <td>-0.231172</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>-0.350991</td>\n",
              "      <td>0.069753</td>\n",
              "      <td>0.560066</td>\n",
              "      <td>0.624356</td>\n",
              "      <td>-1.698114</td>\n",
              "      <td>00002.jpg</td>\n",
              "      <td>0.689658</td>\n",
              "      <td>0.767415</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>0.248237</td>\n",
              "      <td>0.003834</td>\n",
              "      <td>-0.091794</td>\n",
              "      <td>-0.130754</td>\n",
              "      <td>-0.141648</td>\n",
              "      <td>00003.jpg</td>\n",
              "      <td>-0.154704</td>\n",
              "      <td>-0.169938</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>6.803063</td>\n",
              "      <td>2.678256</td>\n",
              "      <td>2.063982</td>\n",
              "      <td>2.490136</td>\n",
              "      <td>0.618897</td>\n",
              "      <td>00004.jpg</td>\n",
              "      <td>0.772491</td>\n",
              "      <td>1.860208</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>-0.620033</td>\n",
              "      <td>-0.683605</td>\n",
              "      <td>-1.234380</td>\n",
              "      <td>-1.164447</td>\n",
              "      <td>0.123658</td>\n",
              "      <td>00005.jpg</td>\n",
              "      <td>-1.215500</td>\n",
              "      <td>-1.154394</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5cbbdd73-8709-41ff-8419-b3893033986f')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-5cbbdd73-8709-41ff-8419-b3893033986f 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-5cbbdd73-8709-41ff-8419-b3893033986f');\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",
              "\n",
              "\n",
              "<div id=\"df-4e1f1998-1c26-4464-93ae-a0183eef29b6\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-4e1f1998-1c26-4464-93ae-a0183eef29b6')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-4e1f1998-1c26-4464-93ae-a0183eef29b6 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "train_data_standardized",
              "summary": "{\n  \"name\": \"train_data_standardized\",\n  \"rows\": 12948,\n  \"fields\": [\n    {\n      \"column\": \"Unnamed: 0\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2200,\n        \"min\": 0,\n        \"max\": 8143,\n        \"num_unique_values\": 8144,\n        \"samples\": [\n          7437,\n          483,\n          7598\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.0000386182394088,\n        \"min\": -0.7790115752093401,\n        \"max\": 19.36240300434475,\n        \"num_unique_values\": 474,\n        \"samples\": [\n          4.968690217377437,\n          4.21048274686599,\n          -0.5344285202056474\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.0000386182394088,\n        \"min\": -1.0131993492330233,\n        \"max\": 14.524814490920646,\n        \"num_unique_values\": 589,\n        \"samples\": [\n          3.5069501315389533,\n          1.7742261639096955,\n          4.07196881663545\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.0000386182394088,\n        \"min\": -1.3735407022675117,\n        \"max\": 11.14851311282172,\n        \"num_unique_values\": 1643,\n        \"samples\": [\n          -0.4995115999469922,\n          -0.6899425216816305,\n          -1.1611369818711843\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"y2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.000038618239409,\n        \"min\": -1.3770510845978332,\n        \"max\": 10.895312033329438,\n        \"num_unique_values\": 1247,\n        \"samples\": [\n          1.2694979509886333,\n          3.171934169090262,\n          0.4667358473966164\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"class\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.0000386182394088,\n        \"min\": -1.733488229277107,\n        \"max\": 1.7154978666891316,\n        \"num_unique_values\": 196,\n        \"samples\": [\n          -1.6273655801704534,\n          -0.8668199282394368,\n          -0.46001643999726505\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"fname\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 12948,\n        \"samples\": [\n          \"07278.jpg\",\n          \"05322.jpg\",\n          \"08421.jpg\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"bbox_width\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.0000386182394088,\n        \"min\": -1.3437572775933153,\n        \"max\": 10.61090709492222,\n        \"num_unique_values\": 1555,\n        \"samples\": [\n          0.8740280217334381,\n          0.05905872499288416,\n          -0.1413435610908586\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"bbox_height\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.0000386182394088,\n        \"min\": -1.3428065412497676,\n        \"max\": 12.185218832696213,\n        \"num_unique_values\": 1017,\n        \"samples\": [\n          2.368921869150453,\n          0.05144677027112742,\n          1.8696284535382406\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Handling text and Categorial Attributes"
      ],
      "metadata": {
        "id": "SQMSHm8wjBwn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "2bztjVLAjBgu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "car_cat = train_data[\"fname\"]\n",
        "car_cat.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6ZXuh_FrjAde",
        "outputId": "20e60afd-6e75-4758-c6e7-2974ad610a0d"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    00001.jpg\n",
              "1    00002.jpg\n",
              "2    00003.jpg\n",
              "3    00004.jpg\n",
              "4    00005.jpg\n",
              "Name: fname, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import OneHotEncoder\n",
        "\n",
        "car_cat_reshaped = car_cat.values.reshape(-1, 1)\n",
        "cat_encoder = OneHotEncoder()\n",
        "\n",
        "car_cat_1hot = cat_encoder.fit_transform(car_cat_reshaped)"
      ],
      "metadata": {
        "id": "Jjy48kCTjJnj"
      },
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "car_cat_1hot"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IgjiEGt7jfCW",
        "outputId": "76eb85f6-c5ff-4c77-b5c8-38c9fcb1e925"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<12948x12948 sparse matrix of type '<class 'numpy.float64'>'\n",
              "\twith 12948 stored elements in Compressed Sparse Row format>"
            ]
          },
          "metadata": {},
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "car_cat_1hot.toarray()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uN59NcwujnsT",
        "outputId": "3bce91c1-9943-4eea-ccec-1b82758fd86e"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1., 0., 0., ..., 0., 0., 0.],\n",
              "       [0., 1., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 1., ..., 0., 0., 0.],\n",
              "       ...,\n",
              "       [0., 0., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 0., ..., 1., 0., 0.],\n",
              "       [0., 0., 0., ..., 0., 1., 0.]])"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "car_cat_1hot.toarray().shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RqUJFbLCjmuN",
        "outputId": "2c36f7bf-720c-4221-e81e-bb8908aa7533"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(12948, 12948)"
            ]
          },
          "metadata": {},
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cat_encoder.categories_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PnthlrgGkFhN",
        "outputId": "76154f48-cb19-4c44-a970-bd8e31f56417"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[array(['00001.jpg', '00002.jpg', '00003.jpg', ..., '12947.jpg',\n",
              "        '12948.jpg', '9999.jpg'], dtype=object)]"
            ]
          },
          "metadata": {},
          "execution_count": 47
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "A1-ue97mkKo4"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}