Laboratory_exercise_4.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Laboratory exercise 4\n",
        "\n",
        "## Warm-Up Mode (2 points)\n",
        "\n",
        "**Task Description**  \n",
        "Using the given dataset, develop and implement **3** different neural networks to predict the **air quality level**. Each network should differ in the following ways:  \n",
        "\n",
        "- **layer configurations** - use different numbers and types of layers;\n",
        "- **activation functions** - try different activation functions;\n",
        "- **neurons per layer** - experiment with different numbers of neurons in each layer; and\n",
        "- **number of layers** - build networks with varying depths.\n",
        "\n",
        "After developing the models, evaluate and compare the performance of all **3** approaches.\n",
        "\n",
        "**About the Dataset**  \n",
        "This dataset focuses on air quality assessment across various regions. The dataset contains 5,000 samples and captures critical environmental and demographic factors that influence pollution levels.\n",
        "\n",
        "**Features**:  \n",
        "- **Temperature (°C)**: Average temperature of the region.  \n",
        "- **Humidity (%)**: Relative humidity recorded in the region.  \n",
        "- **PM2.5 Concentration (µg/m³)**: Levels of fine particulate matter.  \n",
        "- **PM10 Concentration (µg/m³)**: Levels of coarse particulate matter.  \n",
        "- **NO2 Concentration (ppb)**: Nitrogen dioxide levels.  \n",
        "- **SO2 Concentration (ppb)**: Sulfur dioxide levels.  \n",
        "- **CO Concentration (ppm)**: Carbon monoxide levels.  \n",
        "- **Proximity to Industrial Areas (km)**: Distance to the nearest industrial zone.  \n",
        "- **Population Density (people/km²)**: Number of people per square kilometer in the region.  \n",
        "\n",
        "**Target Variable**: **Air Quality**  \n",
        "- **Good**: Clean air with low pollution levels.  \n",
        "- **Moderate**: Acceptable air quality but with some pollutants present.  \n",
        "- **Poor**: Noticeable pollution that may cause health issues for sensitive groups.  \n",
        "- **Hazardous**: Highly polluted air posing serious health risks to the population.  "
      ],
      "metadata": {
        "id": "t7JOxuK9Jje2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Imports**"
      ],
      "metadata": {
        "id": "wOQhg8Qu3yyC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Dropout"
      ],
      "metadata": {
        "id": "KopRF6dztEXJ"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Dataset**"
      ],
      "metadata": {
        "id": "iBR0xKIM4fw4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('./sample_data/pollution_dataset.csv')\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "3zruxnPk4kwp",
        "outputId": "b79b71cd-d0cd-40d9-cf07-f768239624f1"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Temperature  Humidity  PM2.5  PM10   NO2   SO2    CO  \\\n",
              "0         29.8      59.1    5.2  17.9  18.9   9.2  1.72   \n",
              "1         28.3      75.6    2.3  12.2  30.8   9.7  1.64   \n",
              "2         23.1      74.7   26.7  33.8  24.4  12.6  1.63   \n",
              "3         27.1      39.1    6.1   6.3  13.5   5.3  1.15   \n",
              "4         26.5      70.7    6.9  16.0  21.9   5.6  1.01   \n",
              "\n",
              "   Proximity_to_Industrial_Areas  Population_Density Air Quality  \n",
              "0                            6.3                 319    Moderate  \n",
              "1                            6.0                 611    Moderate  \n",
              "2                            5.2                 619    Moderate  \n",
              "3                           11.1                 551        Good  \n",
              "4                           12.7                 303        Good  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-74bd1013-5ebc-441f-a2d0-5667b8a26ac3\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Temperature</th>\n",
              "      <th>Humidity</th>\n",
              "      <th>PM2.5</th>\n",
              "      <th>PM10</th>\n",
              "      <th>NO2</th>\n",
              "      <th>SO2</th>\n",
              "      <th>CO</th>\n",
              "      <th>Proximity_to_Industrial_Areas</th>\n",
              "      <th>Population_Density</th>\n",
              "      <th>Air Quality</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>29.8</td>\n",
              "      <td>59.1</td>\n",
              "      <td>5.2</td>\n",
              "      <td>17.9</td>\n",
              "      <td>18.9</td>\n",
              "      <td>9.2</td>\n",
              "      <td>1.72</td>\n",
              "      <td>6.3</td>\n",
              "      <td>319</td>\n",
              "      <td>Moderate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>28.3</td>\n",
              "      <td>75.6</td>\n",
              "      <td>2.3</td>\n",
              "      <td>12.2</td>\n",
              "      <td>30.8</td>\n",
              "      <td>9.7</td>\n",
              "      <td>1.64</td>\n",
              "      <td>6.0</td>\n",
              "      <td>611</td>\n",
              "      <td>Moderate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>23.1</td>\n",
              "      <td>74.7</td>\n",
              "      <td>26.7</td>\n",
              "      <td>33.8</td>\n",
              "      <td>24.4</td>\n",
              "      <td>12.6</td>\n",
              "      <td>1.63</td>\n",
              "      <td>5.2</td>\n",
              "      <td>619</td>\n",
              "      <td>Moderate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>27.1</td>\n",
              "      <td>39.1</td>\n",
              "      <td>6.1</td>\n",
              "      <td>6.3</td>\n",
              "      <td>13.5</td>\n",
              "      <td>5.3</td>\n",
              "      <td>1.15</td>\n",
              "      <td>11.1</td>\n",
              "      <td>551</td>\n",
              "      <td>Good</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>26.5</td>\n",
              "      <td>70.7</td>\n",
              "      <td>6.9</td>\n",
              "      <td>16.0</td>\n",
              "      <td>21.9</td>\n",
              "      <td>5.6</td>\n",
              "      <td>1.01</td>\n",
              "      <td>12.7</td>\n",
              "      <td>303</td>\n",
              "      <td>Good</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-74bd1013-5ebc-441f-a2d0-5667b8a26ac3')\"\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-74bd1013-5ebc-441f-a2d0-5667b8a26ac3 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-74bd1013-5ebc-441f-a2d0-5667b8a26ac3');\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-9251dd1f-86e0-473c-a860-ec90a1302ec1\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-9251dd1f-86e0-473c-a860-ec90a1302ec1')\"\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-9251dd1f-86e0-473c-a860-ec90a1302ec1 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": "df",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 5000,\n  \"fields\": [\n    {\n      \"column\": \"Temperature\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.72066137351523,\n        \"min\": 13.4,\n        \"max\": 58.6,\n        \"num_unique_values\": 362,\n        \"samples\": [\n          20.3,\n          27.7,\n          32.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Humidity\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 15.863576682407786,\n        \"min\": 36.0,\n        \"max\": 128.1,\n        \"num_unique_values\": 723,\n        \"samples\": [\n          74.3,\n          58.5,\n          86.2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"PM2.5\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 24.554545879838066,\n        \"min\": 0.0,\n        \"max\": 295.0,\n        \"num_unique_values\": 815,\n        \"samples\": [\n          21.7,\n          11.5,\n          52.1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"PM10\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 27.349199378624032,\n        \"min\": -0.2,\n        \"max\": 315.8,\n        \"num_unique_values\": 955,\n        \"samples\": [\n          18.4,\n          97.4,\n          83.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"NO2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 8.895356376391709,\n        \"min\": 7.4,\n        \"max\": 64.9,\n        \"num_unique_values\": 445,\n        \"samples\": [\n          19.9,\n          19.8,\n          33.1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"SO2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.750303377385161,\n        \"min\": -6.2,\n        \"max\": 44.9,\n        \"num_unique_values\": 348,\n        \"samples\": [\n          18.7,\n          13.4,\n          35.7\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"CO\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.5460266694116166,\n        \"min\": 0.65,\n        \"max\": 3.72,\n        \"num_unique_values\": 265,\n        \"samples\": [\n          2.27,\n          2.02,\n          2.97\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Proximity_to_Industrial_Areas\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3.610943730235125,\n        \"min\": 2.5,\n        \"max\": 25.8,\n        \"num_unique_values\": 179,\n        \"samples\": [\n          9.0,\n          4.1,\n          8.2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Population_Density\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 152,\n        \"min\": 188,\n        \"max\": 957,\n        \"num_unique_values\": 683,\n        \"samples\": [\n          295,\n          679,\n          492\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Air Quality\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"Good\",\n          \"Poor\",\n          \"Moderate\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Splitting Features**"
      ],
      "metadata": {
        "id": "Ycv46srQ6MDw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X = df.drop(columns=['Air Quality'])\n",
        "y = df['Air Quality']"
      ],
      "metadata": {
        "id": "Lsc8fnEw6Xei"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Encoding**"
      ],
      "metadata": {
        "id": "82wziVZW6cX_"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The target (Air Quality) has text labels (e.g., \"Good\", \"Moderate\"). Machine learning models can’t work with text, so we convert these into numerical labels (e.g., 0, 1, 2, 3)."
      ],
      "metadata": {
        "id": "yJsHuXCh6jCk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "label_encoder = LabelEncoder()\n",
        "y = label_encoder.fit_transform(y)"
      ],
      "metadata": {
        "id": "8ALy7yVA6eaK"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Splitting**"
      ],
      "metadata": {
        "id": "admHPLFG43C5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
      ],
      "metadata": {
        "id": "E_2d7JNS6ukY"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Scaling Features**"
      ],
      "metadata": {
        "id": "NuxsS-Rq6ygQ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Neural networks perform better when input features are on a similar scale. Features like \"Temperature\" (measured in °C) and \"Population Density\" (people/km²) might have very different ranges, causing the model to focus more on one feature than others."
      ],
      "metadata": {
        "id": "DxisbZPm66a-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "scaler = StandardScaler()\n",
        "X_train = scaler.fit_transform(X_train)\n",
        "X_test = scaler.transform(X_test)"
      ],
      "metadata": {
        "id": "FRqnpA4362I0"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Models**"
      ],
      "metadata": {
        "id": "1cWrWm7lA4DW"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Model 1"
      ],
      "metadata": {
        "id": "OdbOzE--A786"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_1 = Sequential([\n",
        "  Dense(64, activation='relu', input_shape=(X_train.shape[1],)),\n",
        "  Dense(32, activation='relu'),\n",
        "  Dropout(0.3),\n",
        "  Dense(len(np.unique(y)), activation='softmax')\n",
        "])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MJVErlk5A7UF",
        "outputId": "bb85e780-ae83-4bbd-af65-6907f475f0a3"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "_MP0v04ZBDIR"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Model 2"
      ],
      "metadata": {
        "id": "1DzjpA-S67rC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_2 = Sequential([\n",
        "  Dense(128, activation='tanh', input_shape=(X_train.shape[1],)),\n",
        "  Dense(64, activation='tanh'),\n",
        "  Dense(32, activation='tanh'),\n",
        "  Dense(16, activation='tanh'),\n",
        "  Dense(len(np.unique(y)), activation='softmax')\n",
        "])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uAGR2IzJBWYL",
        "outputId": "58d0f23c-ad80-41dd-aff8-6a4924b73a57"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.compile(optimizer='rmsprop', loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "_uGInmkdBkXN"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Model 3"
      ],
      "metadata": {
        "id": "bI7zykgdBmLy"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_3 = Sequential([\n",
        "  Dense(256, activation='sigmoid', input_shape=(X_train.shape[1],)),\n",
        "  Dense(len(np.unique(y)), activation='softmax')\n",
        "])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zliJjBd1Bpzw",
        "outputId": "08244137-5963-4ac7-802b-412bdc0d0839"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3.compile(optimizer='sgd', loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "fqcQy4PYBvqV"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Training**"
      ],
      "metadata": {
        "id": "BoEcSHPABsdq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "history_1 = model_1.fit(X_train, y_train, epochs=20, validation_split=0.2, batch_size=32, verbose=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x3IA0ckPB6qG",
        "outputId": "dd8b0579-874b-4cb5-c6fd-fc48e9f7ab17"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 4ms/step - accuracy: 0.4950 - loss: 1.1267 - val_accuracy: 0.8450 - val_loss: 0.5065\n",
            "Epoch 2/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8121 - loss: 0.4930 - val_accuracy: 0.8938 - val_loss: 0.3102\n",
            "Epoch 3/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8742 - loss: 0.3411 - val_accuracy: 0.9150 - val_loss: 0.2442\n",
            "Epoch 4/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9025 - loss: 0.2628 - val_accuracy: 0.9125 - val_loss: 0.2152\n",
            "Epoch 5/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.8969 - loss: 0.2549 - val_accuracy: 0.9212 - val_loss: 0.2051\n",
            "Epoch 6/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9308 - loss: 0.2028 - val_accuracy: 0.9312 - val_loss: 0.1919\n",
            "Epoch 7/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9215 - loss: 0.2054 - val_accuracy: 0.9287 - val_loss: 0.1833\n",
            "Epoch 8/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9278 - loss: 0.1882 - val_accuracy: 0.9350 - val_loss: 0.1825\n",
            "Epoch 9/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9331 - loss: 0.1841 - val_accuracy: 0.9375 - val_loss: 0.1718\n",
            "Epoch 10/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9317 - loss: 0.1785 - val_accuracy: 0.9325 - val_loss: 0.1754\n",
            "Epoch 11/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9343 - loss: 0.1789 - val_accuracy: 0.9400 - val_loss: 0.1670\n",
            "Epoch 12/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9336 - loss: 0.1756 - val_accuracy: 0.9388 - val_loss: 0.1736\n",
            "Epoch 13/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9380 - loss: 0.1632 - val_accuracy: 0.9400 - val_loss: 0.1616\n",
            "Epoch 14/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9453 - loss: 0.1527 - val_accuracy: 0.9350 - val_loss: 0.1640\n",
            "Epoch 15/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9402 - loss: 0.1630 - val_accuracy: 0.9400 - val_loss: 0.1581\n",
            "Epoch 16/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9487 - loss: 0.1453 - val_accuracy: 0.9362 - val_loss: 0.1691\n",
            "Epoch 17/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9399 - loss: 0.1604 - val_accuracy: 0.9375 - val_loss: 0.1637\n",
            "Epoch 18/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9486 - loss: 0.1373 - val_accuracy: 0.9362 - val_loss: 0.1674\n",
            "Epoch 19/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9479 - loss: 0.1475 - val_accuracy: 0.9375 - val_loss: 0.1616\n",
            "Epoch 20/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9542 - loss: 0.1298 - val_accuracy: 0.9362 - val_loss: 0.1589\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history_2 = model_2.fit(X_train, y_train, epochs=20, validation_split=0.2, batch_size=32, verbose=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KyMhVNLACLMY",
        "outputId": "af827714-fc82-4dca-c2f5-c9e46c01a1b2"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.7832 - loss: 0.6495 - val_accuracy: 0.9325 - val_loss: 0.2764\n",
            "Epoch 2/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9254 - loss: 0.2541 - val_accuracy: 0.9375 - val_loss: 0.2155\n",
            "Epoch 3/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9340 - loss: 0.2084 - val_accuracy: 0.9250 - val_loss: 0.2048\n",
            "Epoch 4/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.9217 - loss: 0.2038 - val_accuracy: 0.9087 - val_loss: 0.2152\n",
            "Epoch 5/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9306 - loss: 0.1820 - val_accuracy: 0.9300 - val_loss: 0.1959\n",
            "Epoch 6/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9293 - loss: 0.1789 - val_accuracy: 0.9337 - val_loss: 0.1732\n",
            "Epoch 7/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9452 - loss: 0.1549 - val_accuracy: 0.8925 - val_loss: 0.2498\n",
            "Epoch 8/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9367 - loss: 0.1666 - val_accuracy: 0.9300 - val_loss: 0.1781\n",
            "Epoch 9/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9368 - loss: 0.1584 - val_accuracy: 0.9275 - val_loss: 0.1866\n",
            "Epoch 10/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9433 - loss: 0.1540 - val_accuracy: 0.9337 - val_loss: 0.1704\n",
            "Epoch 11/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9513 - loss: 0.1407 - val_accuracy: 0.9325 - val_loss: 0.1722\n",
            "Epoch 12/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9389 - loss: 0.1561 - val_accuracy: 0.9325 - val_loss: 0.1710\n",
            "Epoch 13/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9418 - loss: 0.1433 - val_accuracy: 0.9337 - val_loss: 0.1775\n",
            "Epoch 14/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9481 - loss: 0.1445 - val_accuracy: 0.9325 - val_loss: 0.1699\n",
            "Epoch 15/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9492 - loss: 0.1341 - val_accuracy: 0.9388 - val_loss: 0.1612\n",
            "Epoch 16/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.9467 - loss: 0.1347 - val_accuracy: 0.9388 - val_loss: 0.1639\n",
            "Epoch 17/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.9524 - loss: 0.1250 - val_accuracy: 0.9325 - val_loss: 0.1628\n",
            "Epoch 18/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - accuracy: 0.9515 - loss: 0.1224 - val_accuracy: 0.9237 - val_loss: 0.1834\n",
            "Epoch 19/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9531 - loss: 0.1276 - val_accuracy: 0.9287 - val_loss: 0.1787\n",
            "Epoch 20/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.9481 - loss: 0.1276 - val_accuracy: 0.9350 - val_loss: 0.1703\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history_3 = model_3.fit(X_train, y_train, epochs=20, validation_split=0.2, batch_size=32, verbose=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8tLzNMUbCbFB",
        "outputId": "0d275752-72bb-4571-c0fa-8e4a1e028a36"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - accuracy: 0.3938 - loss: 1.2825 - val_accuracy: 0.5000 - val_loss: 1.1034\n",
            "Epoch 2/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4810 - loss: 1.0547 - val_accuracy: 0.5437 - val_loss: 0.9591\n",
            "Epoch 3/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5877 - loss: 0.9233 - val_accuracy: 0.6600 - val_loss: 0.8632\n",
            "Epoch 4/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6737 - loss: 0.8334 - val_accuracy: 0.7150 - val_loss: 0.7844\n",
            "Epoch 5/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.7264 - loss: 0.7714 - val_accuracy: 0.7237 - val_loss: 0.7258\n",
            "Epoch 6/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.7503 - loss: 0.7137 - val_accuracy: 0.7775 - val_loss: 0.6781\n",
            "Epoch 7/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.7815 - loss: 0.6603 - val_accuracy: 0.7487 - val_loss: 0.6469\n",
            "Epoch 8/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.7893 - loss: 0.6258 - val_accuracy: 0.7825 - val_loss: 0.6097\n",
            "Epoch 9/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8168 - loss: 0.5970 - val_accuracy: 0.8200 - val_loss: 0.5739\n",
            "Epoch 10/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8330 - loss: 0.5665 - val_accuracy: 0.8313 - val_loss: 0.5459\n",
            "Epoch 11/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8437 - loss: 0.5308 - val_accuracy: 0.8425 - val_loss: 0.5242\n",
            "Epoch 12/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8560 - loss: 0.5192 - val_accuracy: 0.8537 - val_loss: 0.5003\n",
            "Epoch 13/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8601 - loss: 0.5002 - val_accuracy: 0.8625 - val_loss: 0.4834\n",
            "Epoch 14/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8659 - loss: 0.4791 - val_accuracy: 0.8750 - val_loss: 0.4683\n",
            "Epoch 15/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8759 - loss: 0.4614 - val_accuracy: 0.8737 - val_loss: 0.4496\n",
            "Epoch 16/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8841 - loss: 0.4377 - val_accuracy: 0.8900 - val_loss: 0.4354\n",
            "Epoch 17/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8877 - loss: 0.4324 - val_accuracy: 0.8950 - val_loss: 0.4235\n",
            "Epoch 18/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8898 - loss: 0.4159 - val_accuracy: 0.9000 - val_loss: 0.4145\n",
            "Epoch 19/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8961 - loss: 0.4094 - val_accuracy: 0.9025 - val_loss: 0.3994\n",
            "Epoch 20/20\n",
            "\u001b[1m100/100\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9023 - loss: 0.3859 - val_accuracy: 0.9087 - val_loss: 0.3903\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Comparison**"
      ],
      "metadata": {
        "id": "lty6ISD3Csh2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(history_1.history['accuracy'], label='Model 1')\n",
        "plt.plot(history_2.history['accuracy'], label='Model 2')\n",
        "plt.plot(history_3.history['accuracy'], label='Model 3')\n",
        "plt.legend()\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.title('Training Accuracy Comparison')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "w5ZzMl8rCu6r",
        "outputId": "9cf98903-5d62-4478-bee2-34812d725556"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred_1 = model_1.predict(X_test)\n",
        "y_pred_2 = model_2.predict(X_test)\n",
        "y_pred_3 = model_3.predict(X_test)\n",
        "\n",
        "y_pred_1_classes = np.argmax(y_pred_1, axis=1)\n",
        "y_pred_2_classes = np.argmax(y_pred_2, axis=1)\n",
        "y_pred_3_classes = np.argmax(y_pred_3, axis=1)\n",
        "\n",
        "accuracy_1 = accuracy_score(y_test, y_pred_1_classes)\n",
        "accuracy_2 = accuracy_score(y_test, y_pred_2_classes)\n",
        "accuracy_3 = accuracy_score(y_test, y_pred_3_classes)\n",
        "\n",
        "print(f\"Model 1 - Accuracy: {accuracy_1 * 100:.2f}%\")\n",
        "print(f\"Model 2 - Accuracy: {accuracy_2 * 100:.2f}%\")\n",
        "print(f\"Model 3 - Accuracy: {accuracy_3 * 100:.2f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pmegOcxdDzKJ",
        "outputId": "39626565-00f1-424b-fb86-b547029e173d"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m32/32\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step\n",
            "\u001b[1m32/32\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 14ms/step\n",
            "\u001b[1m32/32\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 11ms/step\n",
            "Model 1 - Accuracy: 95.20%\n",
            "Model 2 - Accuracy: 95.40%\n",
            "Model 3 - Accuracy: 90.50%\n"
          ]
        }
      ]
    }
  ]
}