main.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "5-UK-Qy4QvWJ"
      ]
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Image classification to predict deforestation type\n",
        "\n",
        "Event: [Schneider Electric European Hackaton](https://nuwe.io/dev/event/schneider-electric-european-hackathon)\n",
        "\n",
        "Category: Data Science challenge\n",
        "\n",
        "Challenge: [Zero deforestation mission](https://nuwe.io/dev/challenges/data-science-se-european)\n",
        "\n",
        "Author: [Sergi Salido](https://github.com/SergiSalido)\n",
        "\n",
        "Date: 2022-11-19\n",
        "\n",
        "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](http://colab.research.google.com/github/SergiSalido/SE_Hackaton_DataScience_ZeroDeforestationMission/blob/main/main.ipynb)\n"
      ],
      "metadata": {
        "id": "5-UK-Qy4QvWJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Challenge goal\n",
        "The challenge consists in creating an image classification model, which from a given dataset, **predicts the class of deforestation appears in the image with the objective of early detection of this type of actions in protected lands**.\n",
        "\n",
        "## Tasks\n",
        "1. Create an image classification predictive model\n",
        "1. Train the model with the training images\n",
        "1. Find the model that maximizes the F1-score (macro.)\n",
        "1. Test the model to classify the testing images\n",
        "1. Evaluate the F1-score of the model with predictions\n",
        "\n",
        "## Datasets\n",
        "The following files are provided to complete the challenge:\n",
        "* [train.csv](https://challenges-asset-files.s3.us-east-2.amazonaws.com/data_sets/Data-Science/4+-+events/SchneiderElectricEuropeanHackathon22/train.csv): A training CSV file that relates the ID of the images to their features and classes.\n",
        "* [test.csv](https://challenges-asset-files.s3.us-east-2.amazonaws.com/data_sets/Data-Science/4+-+events/SchneiderElectricEuropeanHackathon22/test.csv): A testing CSV file that relates the ID of the images to their features.\n",
        "* [train_test_data.zip](https://challenges-asset-files.s3.us-east-2.amazonaws.com/data_sets/Data-Science/4+-+events/SchneiderElectricEuropeanHackathon22/train_test_data.zip): A zip file containing a few thousand training and testing aerial images of different deforested areas. This zip file has a weight of 309MB.\n",
        "\n",
        "These are the attributes to be able to make the classifications:\n",
        "\n",
        "* **latitude**: Where the photo latitude was taken.\n",
        "* **longitude**: Where the photo longitude was taken.\n",
        "* **year**: Year, in which the photo was taken.\n",
        "* **example_path**: Path where the sample image is located.\n",
        "* **label**: In this column you will have the following categories:\n",
        "  * **Plantation**: Encoded with number 0, Network of rectangular plantation blocks, connected by a well-defined road grid. In hilly areas the layout of the plantation may follow topographic features. In this group you can find: Oil Palm Plantation, Timber Plantation and Other large-scale plantations.\n",
        "  * **Grassland/Shrubland**: Encoded with number 1, Large homogeneous areas with few or sparse shrubs or trees, and which are generally persistent. Distinguished by the absence of signs of agriculture, such as clearly defined field boundaries.\n",
        "  * **Smallholder Agriculture**: Encoded with number 2, Small scale area, in which you can find deforestation covered by agriculture, mixed plantation or oil palm plantation.\n",
        "\n",
        "The neural network model will predict the class (0, 1, 2) of the deforested area."
      ],
      "metadata": {
        "id": "llahSbQfZMke"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Source code\n",
        "The source code is divided into the following steps:\n",
        "<nav>\n",
        "* [Importing libraries](#1)\n",
        "* [Loading data](#1)\n",
        "* [Exploring data](#1)\n",
        "* [Defining the model structure](#1)\n",
        "* [Training the model](#1)\n",
        "* [Making predictions](#1)\n",
        "</nav>"
      ],
      "metadata": {
        "id": "u8BkM9vjAr4F"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "<a id=\"1\"></a>\n",
        "## Importing libraries"
      ],
      "metadata": {
        "id": "BIHyGLoNj-rn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import pandas\n",
        "import tensorflow as tf\n",
        "import matplotlib.pyplot as plt\n",
        "import datetime\n",
        "\n",
        "from PIL import Image\n",
        "\n",
        "import keras\n",
        "from keras.models import Sequential\n",
        "import keras.layers as layers\n",
        "from keras.utils import to_categorical\n",
        "from keras.preprocessing import image\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from keras.utils import to_categorical\n",
        "from tqdm import tqdm"
      ],
      "metadata": {
        "id": "WM1SZ_hKkZF9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "<a id=\"2\"></a>\n",
        "## Loading data"
      ],
      "metadata": {
        "id": "pjHm4moxSjyR"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "uISlOdrJKCXe"
      },
      "outputs": [],
      "source": [
        "# Read the training dataset\n",
        "train_df = pandas.read_csv('https://raw.githubusercontent.com/SergiSalido/SE_Hackaton_DataScience_ZeroDeforestationMission/main/datasets/train.csv')\n",
        "\n",
        "# Read the testing dataset\n",
        "test_df = pandas.read_csv('https://raw.githubusercontent.com/SergiSalido/SE_Hackaton_DataScience_ZeroDeforestationMission/main/datasets/test.csv')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Download the training and testing images zip file\n",
        "!wget https://challenges-asset-files.s3.us-east-2.amazonaws.com/data_sets/Data-Science/4+-+events/SchneiderElectricEuropeanHackathon22/train_test_data.zip &> /dev/null\n",
        "# Unzip the zip file\n",
        "!unzip train_test_data.zip &> /dev/null\n",
        "# Removes the zip file\n",
        "!rm train_test_data.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aybRndxnE_8J",
        "outputId": "28a8ab8b-3967-4355-8dcd-78c6c9ab3335"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "^C\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Exploring the data\n",
        "Exploratory Data Analysis (EDA) of the given datasets."
      ],
      "metadata": {
        "id": "OUFAs5CgFuCR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Describe the training dataset\n",
        "train_df.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "kp9r30eCwElX",
        "outputId": "3e7685f0-57e1-47d0-f60b-bccc8e6165c8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "             label     latitude    longitude         year\n",
              "count  1714.000000  1714.000000  1714.000000  1714.000000\n",
              "mean      0.882147    -0.458578   108.696539  2009.770712\n",
              "std       0.933952     2.146293     9.136482     4.206081\n",
              "min       0.000000    -8.318462    95.902427  2001.000000\n",
              "25%       0.000000    -1.916629   101.627073  2007.000000\n",
              "50%       0.000000    -0.417759   104.104074  2010.000000\n",
              "75%       2.000000     1.062408   114.274851  2013.000000\n",
              "max       2.000000     5.014370   140.900849  2016.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c5c7a9d2-d53e-4cf6-b768-4f4a1b69c116\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>label</th>\n",
              "      <th>latitude</th>\n",
              "      <th>longitude</th>\n",
              "      <th>year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>1714.000000</td>\n",
              "      <td>1714.000000</td>\n",
              "      <td>1714.000000</td>\n",
              "      <td>1714.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.882147</td>\n",
              "      <td>-0.458578</td>\n",
              "      <td>108.696539</td>\n",
              "      <td>2009.770712</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.933952</td>\n",
              "      <td>2.146293</td>\n",
              "      <td>9.136482</td>\n",
              "      <td>4.206081</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>-8.318462</td>\n",
              "      <td>95.902427</td>\n",
              "      <td>2001.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>-1.916629</td>\n",
              "      <td>101.627073</td>\n",
              "      <td>2007.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>-0.417759</td>\n",
              "      <td>104.104074</td>\n",
              "      <td>2010.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>2.000000</td>\n",
              "      <td>1.062408</td>\n",
              "      <td>114.274851</td>\n",
              "      <td>2013.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>2.000000</td>\n",
              "      <td>5.014370</td>\n",
              "      <td>140.900849</td>\n",
              "      <td>2016.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c5c7a9d2-d53e-4cf6-b768-4f4a1b69c116')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-c5c7a9d2-d53e-4cf6-b768-4f4a1b69c116 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-c5c7a9d2-d53e-4cf6-b768-4f4a1b69c116');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Describe the testing dataset\n",
        "test_df.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "GvQMwy39wM0-",
        "outputId": "379179e4-b3a8-4d9e-bbeb-09e3911d9d08"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         latitude   longitude         year\n",
              "count  635.000000  635.000000   635.000000\n",
              "mean    -0.359479  108.772788  2009.617323\n",
              "std      2.247732    9.153662     4.473548\n",
              "min     -8.880926   95.757681  2001.000000\n",
              "25%     -1.792243  101.609698  2006.000000\n",
              "50%     -0.204938  104.103571  2010.000000\n",
              "75%      1.133431  114.849800  2013.000000\n",
              "max      5.152187  140.987545  2016.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-428c6767-5b67-4976-94ad-bd5a8d5c2c4d\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>latitude</th>\n",
              "      <th>longitude</th>\n",
              "      <th>year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>635.000000</td>\n",
              "      <td>635.000000</td>\n",
              "      <td>635.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>-0.359479</td>\n",
              "      <td>108.772788</td>\n",
              "      <td>2009.617323</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>2.247732</td>\n",
              "      <td>9.153662</td>\n",
              "      <td>4.473548</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-8.880926</td>\n",
              "      <td>95.757681</td>\n",
              "      <td>2001.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>-1.792243</td>\n",
              "      <td>101.609698</td>\n",
              "      <td>2006.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>-0.204938</td>\n",
              "      <td>104.103571</td>\n",
              "      <td>2010.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>1.133431</td>\n",
              "      <td>114.849800</td>\n",
              "      <td>2013.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>5.152187</td>\n",
              "      <td>140.987545</td>\n",
              "      <td>2016.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-428c6767-5b67-4976-94ad-bd5a8d5c2c4d')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-428c6767-5b67-4976-94ad-bd5a8d5c2c4d 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-428c6767-5b67-4976-94ad-bd5a8d5c2c4d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df['label'].value_counts().plot(kind='bar', xlabel=\"label\", ylabel=\"Number of training images\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 294
        },
        "id": "W2_W2W9T4Y_-",
        "outputId": "38bb87bf-63f2-42ca-a4b6-2f05d29b4279"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb84dae3d50>"
            ]
          },
          "metadata": {},
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.groupby(['year', 'label']).size().unstack().plot(kind='bar', stacked=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "LUshr4akDJDV",
        "outputId": "7c8ef67a-536f-448b-b8f7-f7f4e50aa759"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb84bcb7490>"
            ]
          },
          "metadata": {},
          "execution_count": 17
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def check_same_image_size(directory, img_height, img_width):\n",
        "  for file in os.listdir(directory):\n",
        "    img = Image.open(directory + file)\n",
        "    if img.height != img_height or img_width != img_width:\n",
        "      return False\n",
        "  return True\n",
        "\n",
        "img_height = 332\n",
        "img_width  = 332\n",
        "\n",
        "print(check_same_image_size('train_test_data/train/', img_height, img_width))\n",
        "print(check_same_image_size('train_test_data/test/', img_height, img_width))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0dEMcJ9rieGX",
        "outputId": "1c620903-0a19-4db8-e2ff-7b4391a26dbf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "True\n",
            "True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(tf.keras.utils.load_img(train_df['example_path'][0], target_size=(128,128)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "FIWUvs3MRlog",
        "outputId": "f5133978-0238-4558-a7cd-f23eaafe17f8"
      },
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fb7c6550490>"
            ]
          },
          "metadata": {},
          "execution_count": 108
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Preprocessing data"
      ],
      "metadata": {
        "id": "kEJDk9k80iCE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Convert images\n",
        "def convert_images(df, target_size_img_height, target_size_img_width):\n",
        "  images = []\n",
        "  for ind in df.index:\n",
        "    img = tf.keras.utils.load_img(df['example_path'][ind], target_size=(target_size_img_height, target_size_img_width, 3))\n",
        "    img = tf.keras.utils.img_to_array(img)\n",
        "    img = img / 255\n",
        "    images.append(img)\n",
        "  return np.array(images)"
      ],
      "metadata": {
        "id": "AAzQPjXPrLx7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "target_size_img_height  = 128\n",
        "target_size_img_width   = 128\n",
        "\n",
        "X = convert_images(train_df, target_size_img_height, target_size_img_width)"
      ],
      "metadata": {
        "id": "JxZr6REK6Mbc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# As it is a multi-class classification problem (3 classes), one-hot encode the target variable\n",
        "y = to_categorical(train_df['label'].values)"
      ],
      "metadata": {
        "id": "mJ0-69caumtU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating a validation set from the training data\n",
        "X_train, X_validation, y_train, y_validation = train_test_split(X, y, random_state=42, test_size=0.2)"
      ],
      "metadata": {
        "id": "IGGVEVZYvKPD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Defining the model’s architecture\n",
        "CNN Image Classification\n"
      ],
      "metadata": {
        "id": "BAgzCLM8-qRX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "num_classes = 3\n",
        "\n",
        "# Image data augmentation\n",
        "data_augmentation = keras.Sequential(\n",
        "  [layers.RandomFlip(\"horizontal\", input_shape=(target_size_img_height, target_size_img_width, 3)),\n",
        "    layers.RandomRotation(0.1),\n",
        "    layers.RandomZoom(0.1)]\n",
        ")\n",
        "\n",
        "# Create the model architecture\n",
        "model = Sequential()\n",
        "model.add(layers.Conv2D(64, (3,3), activation = 'relu', input_shape = (128, 128, 3)))\n",
        "model.add(layers.Conv2D(32, (3,3), activation = 'relu'))\n",
        "model.add(layers.MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(layers.Dropout(0.2))\n",
        "model.add(layers.Flatten())\n",
        "model.add(layers.Dense(num_classes, activation='softmax'))\n",
        "\n",
        "# Compiling the CNN\n",
        "model.compile(optimizer = 'adam',\n",
        "              loss = 'categorical_crossentropy',\n",
        "              metrics = ['accuracy'])\n",
        "\n",
        "display(model.summary())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 436
        },
        "id": "XxxLHxUvTrPh",
        "outputId": "2bd20eee-b5e8-4ea0-ecaa-fa432c742576"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_13\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_24 (Conv2D)          (None, 126, 126, 64)      1792      \n",
            "                                                                 \n",
            " conv2d_25 (Conv2D)          (None, 124, 124, 32)      18464     \n",
            "                                                                 \n",
            " max_pooling2d_12 (MaxPoolin  (None, 62, 62, 32)       0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_12 (Dropout)        (None, 62, 62, 32)        0         \n",
            "                                                                 \n",
            " flatten_6 (Flatten)         (None, 123008)            0         \n",
            "                                                                 \n",
            " dense_6 (Dense)             (None, 3)                 369027    \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 389,283\n",
            "Trainable params: 389,283\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "None"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Training the model"
      ],
      "metadata": {
        "id": "jPe-8xcE-wS4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Train the model on the training set images and validate it using the validation set\n",
        "epochs = 5\n",
        "history = model.fit(X_train, y_train,\n",
        "                    validation_data = (X_validation, y_validation),\n",
        "                    epochs = epochs)\n",
        "\n",
        "acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs_range = range(epochs)\n",
        "\n",
        "plt.figure(figsize=(8, 8))\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
        "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
        "plt.legend(loc='lower right')\n",
        "plt.title('Training and Validation Accuracy')\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(epochs_range, loss, label='Training Loss')\n",
        "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
        "plt.legend(loc='upper right')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 701
        },
        "id": "fE_YAjt0F1yT",
        "outputId": "7981502a-d44e-440a-8a9a-ed9e41390914"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n",
            "43/43 [==============================] - 76s 2s/step - loss: 0.9468 - accuracy: 0.4887 - val_loss: 0.9577 - val_accuracy: 0.5423\n",
            "Epoch 2/5\n",
            "43/43 [==============================] - 83s 2s/step - loss: 0.9223 - accuracy: 0.5536 - val_loss: 1.0421 - val_accuracy: 0.5102\n",
            "Epoch 3/5\n",
            "43/43 [==============================] - 77s 2s/step - loss: 0.8395 - accuracy: 0.6054 - val_loss: 0.8629 - val_accuracy: 0.5831\n",
            "Epoch 4/5\n",
            "43/43 [==============================] - 74s 2s/step - loss: 0.6983 - accuracy: 0.6922 - val_loss: 0.9746 - val_accuracy: 0.5685\n",
            "Epoch 5/5\n",
            "43/43 [==============================] - 73s 2s/step - loss: 0.5321 - accuracy: 0.8067 - val_loss: 0.8780 - val_accuracy: 0.6035\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x576 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Evaluating the model"
      ],
      "metadata": {
        "id": "n_5_CvLz_Ttk"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "-l6bC1gvIi0H"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Testing the model"
      ],
      "metadata": {
        "id": "oNDhpRhK-znv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_test = convert_images(test_df, target_size_img_height, target_size_img_width)"
      ],
      "metadata": {
        "id": "ziDNb-nZB1gB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y_predict = model.predict(X_test) \n",
        "test_classes = np.argmax(y_predict, axis=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SsDKoXlgJatx",
        "outputId": "cf537006-8c33-4c45-8573-b978ab0fdb79"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "20/20 [==============================] - 7s 348ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_classes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "60uTzY9XJ0FY",
        "outputId": "c58b20ff-9da6-4b62-aa80-81cb5f798fcc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 2, 0, 2, 2, 2, 0, 2, 0, 2, 2, 2, 2, 2, 2, 0, 2, 0, 0, 0, 2, 2,\n",
              "       0, 0, 2, 0, 2, 0, 0, 0, 2, 0, 2, 0, 0, 0, 2, 0, 0, 0, 2, 2, 2, 0,\n",
              "       0, 0, 2, 2, 0, 0, 0, 2, 0, 0, 2, 2, 2, 0, 0, 0, 0, 0, 0, 2, 0, 2,\n",
              "       2, 0, 0, 1, 2, 2, 0, 0, 0, 0, 2, 0, 0, 0, 0, 2, 2, 0, 0, 0, 2, 0,\n",
              "       0, 0, 2, 2, 2, 0, 0, 0, 2, 0, 0, 0, 2, 2, 0, 0, 0, 0, 0, 0, 0, 0,\n",
              "       0, 2, 0, 0, 0, 2, 2, 0, 1, 2, 0, 0, 0, 2, 0, 0, 2, 0, 0, 0, 0, 0,\n",
              "       0, 2, 2, 0, 2, 0, 0, 2, 0, 2, 0, 0, 2, 0, 2, 0, 0, 2, 0, 2, 0, 2,\n",
              "       0, 2, 2, 0, 0, 2, 2, 2, 2, 0, 2, 2, 0, 0, 2, 0, 2, 0, 0, 2, 0, 2,\n",
              "       0, 0, 2, 0, 1, 0, 0, 2, 2, 2, 0, 2, 0, 2, 0, 2, 0, 0, 0, 2, 2, 0,\n",
              "       0, 2, 1, 2, 0, 0, 2, 0, 0, 0, 2, 0, 2, 0, 2, 0, 0, 0, 2, 0, 0, 0,\n",
              "       0, 2, 2, 0, 0, 2, 0, 0, 2, 2, 0, 2, 0, 0, 2, 2, 2, 0, 1, 0, 0, 0,\n",
              "       0, 0, 0, 0, 0, 0, 2, 0, 0, 2, 0, 2, 0, 0, 2, 0, 0, 0, 2, 0, 0, 0,\n",
              "       0, 2, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
              "       2, 2, 0, 0, 2, 2, 0, 2, 2, 2, 0, 2, 0, 0, 0, 0, 0, 2, 2, 2, 0, 0,\n",
              "       0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 2, 2, 2, 0, 0, 0, 1, 0, 0, 2, 2,\n",
              "       0, 0, 0, 0, 0, 0, 2, 0, 1, 0, 0, 0, 0, 2, 0, 0, 2, 0, 2, 0, 0, 2,\n",
              "       2, 2, 0, 2, 2, 2, 2, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 2, 0, 0, 0, 0,\n",
              "       0, 0, 2, 2, 0, 0, 0, 0, 0, 0, 0, 2, 2, 0, 0, 0, 2, 0, 0, 2, 0, 2,\n",
              "       0, 2, 0, 0, 0, 0, 0, 2, 0, 0, 2, 0, 0, 0, 2, 0, 2, 0, 2, 0, 0, 1,\n",
              "       0, 0, 0, 2, 0, 0, 2, 0, 2, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0,\n",
              "       0, 2, 0, 0, 2, 2, 2, 0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 0, 2, 2, 0, 0,\n",
              "       1, 0, 0, 0, 0, 2, 0, 2, 2, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 2, 2,\n",
              "       0, 0, 2, 2, 2, 2, 0, 0, 2, 2, 0, 0, 0, 0, 0, 0, 2, 0, 0, 2, 0, 0,\n",
              "       2, 0, 0, 2, 2, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 2, 2, 0, 0, 0, 2, 2,\n",
              "       2, 0, 0, 0, 2, 0, 0, 2, 2, 2, 0, 0, 2, 2, 0, 0, 2, 0, 0, 2, 0, 0,\n",
              "       0, 0, 1, 0, 2, 0, 0, 2, 0, 0, 0, 2, 2, 2, 0, 2, 0, 0, 2, 0, 2, 0,\n",
              "       2, 2, 0, 2, 2, 0, 2, 0, 0, 2, 0, 2, 0, 2, 0, 2, 0, 0, 2, 0, 0, 0,\n",
              "       2, 0, 2, 2, 0, 0, 2, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 2, 2, 0,\n",
              "       2, 0, 2, 0, 0, 0, 0, 2, 2, 2, 2, 0, 0, 0, 2, 2, 2, 0, 0])"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions_df = test_df\n",
        "predictions_df['target'] = test_classes\n",
        "predictions_df['name'] = predictions_df['example_path'].str.replace(\"train_test_data/test/\", \"\").str.replace(\".png\", \"\")\n",
        "predictions_df = predictions_df[['target', 'name']]\n",
        "predictions_df"
      ],
      "metadata": {
        "id": "Kr2BxvH_HGXI",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a96e6b29-8721-4e1f-840f-4d2782d8a48a"
      },
      "execution_count": 109,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:3: FutureWarning: The default value of regex will change from True to False in a future version.\n",
            "  This is separate from the ipykernel package so we can avoid doing imports until\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions_df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "rIALpd9SQBoP",
        "outputId": "620cbfd0-2732-4713-91ed-709a1df9ada7"
      },
      "execution_count": 110,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     target  name\n",
              "0         0    69\n",
              "1         2   469\n",
              "2         0     6\n",
              "3         2   351\n",
              "4         2  1001\n",
              "..      ...   ...\n",
              "630       2   968\n",
              "631       2    54\n",
              "632       2   494\n",
              "633       0   287\n",
              "634       0   603\n",
              "\n",
              "[635 rows x 2 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-53029b17-a342-422e-9c05-0c3cc4f91b60\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>target</th>\n",
              "      <th>name</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>69</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>469</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2</td>\n",
              "      <td>351</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2</td>\n",
              "      <td>1001</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>630</th>\n",
              "      <td>2</td>\n",
              "      <td>968</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>631</th>\n",
              "      <td>2</td>\n",
              "      <td>54</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>632</th>\n",
              "      <td>2</td>\n",
              "      <td>494</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>633</th>\n",
              "      <td>0</td>\n",
              "      <td>287</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>634</th>\n",
              "      <td>0</td>\n",
              "      <td>603</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>635 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-53029b17-a342-422e-9c05-0c3cc4f91b60')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-53029b17-a342-422e-9c05-0c3cc4f91b60 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-53029b17-a342-422e-9c05-0c3cc4f91b60');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 110
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Save results"
      ],
      "metadata": {
        "id": "rCQbcql0Gsiz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "predictions_df.to_csv('predictions.csv' )"
      ],
      "metadata": {
        "id": "sJfQkg3fW2Nw"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "predictions_df = predictions_df.set_index('name')\n",
        "predictions_df.to_json('predictions.json')"
      ],
      "metadata": {
        "id": "PXuKFH2-WkYf"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}