Final_Project_Food.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "izCIaq-dZKeb",
        "7XcjRtMwil2P"
      ],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "**CIS 545 - Rohan Chhaya, Elaine Fan, Kenny Liou**"
      ],
      "metadata": {
        "id": "RUVLuG6wy4iR"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **0.** Introduction"
      ],
      "metadata": {
        "id": "izCIaq-dZKeb"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "For our project, we wanted to explore Open Food Facts, a Kaggle dataset. Specifically, our project has several goals. First, we wanted to conduct some data cleaning and do some descriptive analysis, such as looking at the top countries with the highest nutrition scores. We also looked at a breakdown of the ingredients list and the most common ingredients we can see today from products around the world. Additionally, we looked at the companies that are selling the most products on the market. We also took a glimpse at the nutrition information on the labels and compared different nutrition labels. We summarized our findings in various visualizations, such as boxplots, heat maps, wordclouds, and pie charts. \n",
        "\n",
        "Next, we looked at building supervised machine learning models (both regression and classification), to see if we could predict the health scores and general \"healthiness\" of various products based on relevant features. This included a variety of linear regression and logistic regression (unregularized and regularized), PCA, Random Forests, and Neural Networks. \n",
        "\n",
        "Let's start with uploading the Kaggle Dataset."
      ],
      "metadata": {
        "id": "ECRiMXbFZPhJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Uploading libraries and accessing Kaggle Dataset"
      ],
      "metadata": {
        "id": "Q6nLmlO5ihYS"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "N0ffHmPlheBL"
      },
      "outputs": [],
      "source": [
        "# Sklearn and Pandas Setup\n",
        "import json\n",
        "import glob\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import datetime as dt\n",
        "import re\n",
        "import os\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from matplotlib import cm\n",
        "from google.colab import drive"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **1.** Downloading the Kaggle Dataset"
      ],
      "metadata": {
        "id": "7XcjRtMwil2P"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We are using TSV for this project, `en.openfoodfacts.org.products.tsv` from a Kaggle dataset. The dataset contains information about various food products.\n",
        "\n",
        "To get the data in here:\n",
        "1. Go to this [Kaggle link](https://www.kaggle.com/datasets/arianazmoudeh/airbnbopendata) and create a Kaggle account (unless you already have an account)\n",
        "2. Go to Account and click on \"Create New API Token\" to get the API key in the form of a json file `kaggle.json`\n",
        "3. Upload the `kaggle.json` file to the default location in your Google Drive (Please **DO NOT** upload the json file into any _specific_ folder as it will be difficult for us to debug issues if you deviate from these instructions!).\n",
        "\n",
        "This can be helpful for your project if you decide to use Kaggle."
      ],
      "metadata": {
        "id": "CPdDQpf0jjZN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Run this cell to mount your drive (you will be prompted to sign in)\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "kh-27q8Lin5z",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7aef56b5-2c84-4107-f156-5581b38e1256"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install kaggle"
      ],
      "metadata": {
        "id": "MnJXYYOOisSs",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e03ca634-d5cc-4d82-ab78-3c94d20c5755"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.8/dist-packages (1.5.12)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.8/dist-packages (from kaggle) (2.8.2)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.8/dist-packages (from kaggle) (7.0.0)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.8/dist-packages (from kaggle) (4.64.1)\n",
            "Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.8/dist-packages (from kaggle) (1.15.0)\n",
            "Requirement already satisfied: urllib3 in /usr/local/lib/python3.8/dist-packages (from kaggle) (1.24.3)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.8/dist-packages (from kaggle) (2.25.1)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.8/dist-packages (from kaggle) (2022.12.7)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.8/dist-packages (from python-slugify->kaggle) (1.3)\n",
            "Requirement already satisfied: chardet<5,>=3.0.2 in /usr/local/lib/python3.8/dist-packages (from requests->kaggle) (4.0.0)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.8/dist-packages (from requests->kaggle) (2.10)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Create the kaggle directory and read the uploaded kaggle.json file\n",
        "# (NOTE: Do NOT run this cell more than once unless restarting kernel)\n",
        "!mkdir ~/.kaggle"
      ],
      "metadata": {
        "id": "W0rAI3Tqiuiq"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Read the uploaded kaggle.json file\n",
        "!cp /content/drive/MyDrive/kaggle.json ~/.kaggle/"
      ],
      "metadata": {
        "id": "tJ5FCdpZiw17"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Download dataset\n",
        "!!kaggle datasets download -d openfoodfacts/world-food-facts"
      ],
      "metadata": {
        "id": "CIJWQOQWi0ey",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "93aea115-1ee0-4145-c496-3c93990a0def"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['Downloading world-food-facts.zip to /content',\n",
              " '',\n",
              " '  0% 0.00/109M [00:00<?, ?B/s]',\n",
              " '  5% 5.00M/109M [00:00<00:03, 29.4MB/s]',\n",
              " ' 16% 17.0M/109M [00:00<00:01, 51.2MB/s]',\n",
              " ' 30% 33.0M/109M [00:00<00:01, 70.8MB/s]',\n",
              " ' 38% 41.0M/109M [00:00<00:01, 65.6MB/s]',\n",
              " ' 50% 54.0M/109M [00:00<00:00, 83.0MB/s]',\n",
              " ' 59% 64.0M/109M [00:00<00:00, 88.4MB/s]',\n",
              " ' 68% 74.0M/109M [00:01<00:00, 62.7MB/s]',\n",
              " ' 82% 89.0M/109M [00:01<00:00, 81.7MB/s]',\n",
              " ' 91% 99.0M/109M [00:01<00:00, 80.8MB/s]',\n",
              " ' 99% 108M/109M [00:01<00:00, 67.5MB/s] ',\n",
              " '',\n",
              " '100% 109M/109M [00:01<00:00, 70.3MB/s]']"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Unzip folder in Colab content folder\n",
        "!unzip /content/world-food-facts.zip"
      ],
      "metadata": {
        "id": "jU5DZ6gzjWMj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "03b1848a-b75e-455e-ab13-8a66b5dedd4c"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  /content/world-food-facts.zip\n",
            "  inflating: en.openfoodfacts.org.products.tsv  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Looking at the dataset"
      ],
      "metadata": {
        "id": "39bdnIfBjby3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **2** Data Cleaning"
      ],
      "metadata": {
        "id": "AFXhyYomqMNu"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's read in the TSV file and look at some of the rows."
      ],
      "metadata": {
        "id": "RnIQnwBi6O8n"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Read the tsv file and save it to a dataframe called `df_open_food`\n",
        "df_open_food = pd.read_csv(\"en.openfoodfacts.org.products.tsv\", sep='\\t')\n",
        "\n",
        "# Check out the first five rows\n",
        "df_open_food.head(5)"
      ],
      "metadata": {
        "id": "iwL70wRVjZIq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "7a2dfe85-51ec-40f3-dfd8-762dda4b7aa0"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.8/dist-packages/IPython/core/interactiveshell.py:3326: DtypeWarning: Columns (0,3,5,19,20,24,25,26,27,28,36,37,38,39,48) have mixed types.Specify dtype option on import or set low_memory=False.\n",
            "  exec(code_obj, self.user_global_ns, self.user_ns)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    code                                                url  \\\n",
              "0   3087  http://world-en.openfoodfacts.org/product/0000...   \n",
              "1   4530  http://world-en.openfoodfacts.org/product/0000...   \n",
              "2   4559  http://world-en.openfoodfacts.org/product/0000...   \n",
              "3  16087  http://world-en.openfoodfacts.org/product/0000...   \n",
              "4  16094  http://world-en.openfoodfacts.org/product/0000...   \n",
              "\n",
              "                      creator   created_t      created_datetime  \\\n",
              "0  openfoodfacts-contributors  1474103866  2016-09-17T09:17:46Z   \n",
              "1             usda-ndb-import  1489069957  2017-03-09T14:32:37Z   \n",
              "2             usda-ndb-import  1489069957  2017-03-09T14:32:37Z   \n",
              "3             usda-ndb-import  1489055731  2017-03-09T10:35:31Z   \n",
              "4             usda-ndb-import  1489055653  2017-03-09T10:34:13Z   \n",
              "\n",
              "  last_modified_t last_modified_datetime                    product_name  \\\n",
              "0      1474103893   2016-09-17T09:18:13Z              Farine de blé noir   \n",
              "1      1489069957   2017-03-09T14:32:37Z  Banana Chips Sweetened (Whole)   \n",
              "2      1489069957   2017-03-09T14:32:37Z                         Peanuts   \n",
              "3      1489055731   2017-03-09T10:35:31Z          Organic Salted Nut Mix   \n",
              "4      1489055653   2017-03-09T10:34:13Z                 Organic Polenta   \n",
              "\n",
              "  generic_name quantity  ... fruits-vegetables-nuts_100g  \\\n",
              "0          NaN      1kg  ...                         NaN   \n",
              "1          NaN      NaN  ...                         NaN   \n",
              "2          NaN      NaN  ...                         NaN   \n",
              "3          NaN      NaN  ...                         NaN   \n",
              "4          NaN      NaN  ...                         NaN   \n",
              "\n",
              "  fruits-vegetables-nuts-estimate_100g collagen-meat-protein-ratio_100g  \\\n",
              "0                                  NaN                              NaN   \n",
              "1                                  NaN                              NaN   \n",
              "2                                  NaN                              NaN   \n",
              "3                                  NaN                              NaN   \n",
              "4                                  NaN                              NaN   \n",
              "\n",
              "  cocoa_100g chlorophyl_100g carbon-footprint_100g nutrition-score-fr_100g  \\\n",
              "0        NaN             NaN                   NaN                     NaN   \n",
              "1        NaN             NaN                   NaN                    14.0   \n",
              "2        NaN             NaN                   NaN                     0.0   \n",
              "3        NaN             NaN                   NaN                    12.0   \n",
              "4        NaN             NaN                   NaN                     NaN   \n",
              "\n",
              "  nutrition-score-uk_100g glycemic-index_100g water-hardness_100g  \n",
              "0                     NaN                 NaN                 NaN  \n",
              "1                    14.0                 NaN                 NaN  \n",
              "2                     0.0                 NaN                 NaN  \n",
              "3                    12.0                 NaN                 NaN  \n",
              "4                     NaN                 NaN                 NaN  \n",
              "\n",
              "[5 rows x 163 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7897a9a6-2525-44ab-a5ed-90ef6ad83fc1\">\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>code</th>\n",
              "      <th>url</th>\n",
              "      <th>creator</th>\n",
              "      <th>created_t</th>\n",
              "      <th>created_datetime</th>\n",
              "      <th>last_modified_t</th>\n",
              "      <th>last_modified_datetime</th>\n",
              "      <th>product_name</th>\n",
              "      <th>generic_name</th>\n",
              "      <th>quantity</th>\n",
              "      <th>...</th>\n",
              "      <th>fruits-vegetables-nuts_100g</th>\n",
              "      <th>fruits-vegetables-nuts-estimate_100g</th>\n",
              "      <th>collagen-meat-protein-ratio_100g</th>\n",
              "      <th>cocoa_100g</th>\n",
              "      <th>chlorophyl_100g</th>\n",
              "      <th>carbon-footprint_100g</th>\n",
              "      <th>nutrition-score-fr_100g</th>\n",
              "      <th>nutrition-score-uk_100g</th>\n",
              "      <th>glycemic-index_100g</th>\n",
              "      <th>water-hardness_100g</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3087</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>openfoodfacts-contributors</td>\n",
              "      <td>1474103866</td>\n",
              "      <td>2016-09-17T09:17:46Z</td>\n",
              "      <td>1474103893</td>\n",
              "      <td>2016-09-17T09:18:13Z</td>\n",
              "      <td>Farine de blé noir</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1kg</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4530</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>Banana Chips Sweetened (Whole)</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>14.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4559</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>Peanuts</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>16087</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055731</td>\n",
              "      <td>2017-03-09T10:35:31Z</td>\n",
              "      <td>1489055731</td>\n",
              "      <td>2017-03-09T10:35:31Z</td>\n",
              "      <td>Organic Salted Nut Mix</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>12.0</td>\n",
              "      <td>12.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>16094</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055653</td>\n",
              "      <td>2017-03-09T10:34:13Z</td>\n",
              "      <td>1489055653</td>\n",
              "      <td>2017-03-09T10:34:13Z</td>\n",
              "      <td>Organic Polenta</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 163 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7897a9a6-2525-44ab-a5ed-90ef6ad83fc1')\"\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-7897a9a6-2525-44ab-a5ed-90ef6ad83fc1 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-7897a9a6-2525-44ab-a5ed-90ef6ad83fc1');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We note a lot of NaN values, so let's remove columns with more than 50% duplicates and check for duplicates and data types of each column as well. "
      ],
      "metadata": {
        "id": "dwQa_UqD6S7N"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#TO-DO: Store the number of duplicate rows in total_duplicates\n",
        "total_duplicates = len(df_open_food)-len(df_open_food.drop_duplicates())\n",
        "total_duplicates"
      ],
      "metadata": {
        "id": "AAn9mxJojbLg",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d7c93f8a-b0e8-4c64-d83e-3c893915139e"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#drop columns that have more than 50% of the values as NaN\n",
        "df_open_food = df_open_food.drop(df_open_food.columns[df_open_food.isnull().sum(axis=0) > (0.5*len(df_open_food))], axis=1)\n",
        "df_open_food.head(10)"
      ],
      "metadata": {
        "id": "umWQUhILpeJX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 393
        },
        "outputId": "05ea48cf-822d-4662-a903-28f3e659ad7d"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    code                                                url  \\\n",
              "0   3087  http://world-en.openfoodfacts.org/product/0000...   \n",
              "1   4530  http://world-en.openfoodfacts.org/product/0000...   \n",
              "2   4559  http://world-en.openfoodfacts.org/product/0000...   \n",
              "3  16087  http://world-en.openfoodfacts.org/product/0000...   \n",
              "4  16094  http://world-en.openfoodfacts.org/product/0000...   \n",
              "5  16100  http://world-en.openfoodfacts.org/product/0000...   \n",
              "6  16117  http://world-en.openfoodfacts.org/product/0000...   \n",
              "7  16124  http://world-en.openfoodfacts.org/product/0000...   \n",
              "8  16193  http://world-en.openfoodfacts.org/product/0000...   \n",
              "9  16513  http://world-en.openfoodfacts.org/product/0000...   \n",
              "\n",
              "                      creator   created_t      created_datetime  \\\n",
              "0  openfoodfacts-contributors  1474103866  2016-09-17T09:17:46Z   \n",
              "1             usda-ndb-import  1489069957  2017-03-09T14:32:37Z   \n",
              "2             usda-ndb-import  1489069957  2017-03-09T14:32:37Z   \n",
              "3             usda-ndb-import  1489055731  2017-03-09T10:35:31Z   \n",
              "4             usda-ndb-import  1489055653  2017-03-09T10:34:13Z   \n",
              "5             usda-ndb-import  1489055651  2017-03-09T10:34:11Z   \n",
              "6             usda-ndb-import  1489055730  2017-03-09T10:35:30Z   \n",
              "7             usda-ndb-import  1489055711  2017-03-09T10:35:11Z   \n",
              "8             usda-ndb-import  1489055651  2017-03-09T10:34:11Z   \n",
              "9             usda-ndb-import  1489055654  2017-03-09T10:34:14Z   \n",
              "\n",
              "  last_modified_t last_modified_datetime                       product_name  \\\n",
              "0      1474103893   2016-09-17T09:18:13Z                 Farine de blé noir   \n",
              "1      1489069957   2017-03-09T14:32:37Z     Banana Chips Sweetened (Whole)   \n",
              "2      1489069957   2017-03-09T14:32:37Z                            Peanuts   \n",
              "3      1489055731   2017-03-09T10:35:31Z             Organic Salted Nut Mix   \n",
              "4      1489055653   2017-03-09T10:34:13Z                    Organic Polenta   \n",
              "5      1489055651   2017-03-09T10:34:11Z  Breadshop Honey Gone Nuts Granola   \n",
              "6      1489055730   2017-03-09T10:35:30Z      Organic Long Grain White Rice   \n",
              "7      1489055712   2017-03-09T10:35:12Z                     Organic Muesli   \n",
              "8      1489055651   2017-03-09T10:34:11Z       Organic Dark Chocolate Minis   \n",
              "9      1489055654   2017-03-09T10:34:14Z              Organic Sunflower Oil   \n",
              "\n",
              "                 brands           brands_tags  ... fat_100g  \\\n",
              "0       Ferme t'y R'nao       ferme-t-y-r-nao  ...      NaN   \n",
              "1                   NaN                   NaN  ...    28.57   \n",
              "2        Torn & Glasser          torn-glasser  ...    17.86   \n",
              "3             Grizzlies             grizzlies  ...    57.14   \n",
              "4        Bob's Red Mill        bob-s-red-mill  ...     1.43   \n",
              "5                  Unfi                  unfi  ...    18.27   \n",
              "6              Lundberg              lundberg  ...      NaN   \n",
              "7        Daddy's Muesli        daddy-s-muesli  ...    18.75   \n",
              "8        Equal Exchange        equal-exchange  ...    37.50   \n",
              "9  Napa Valley Naturals  napa-valley-naturals  ...   100.00   \n",
              "\n",
              "  saturated-fat_100g carbohydrates_100g sugars_100g fiber_100g  proteins_100g  \\\n",
              "0                NaN                NaN         NaN        NaN            NaN   \n",
              "1              28.57              64.29       14.29        3.6           3.57   \n",
              "2               0.00              60.71       17.86        7.1          17.86   \n",
              "3               5.36              17.86        3.57        7.1          17.86   \n",
              "4                NaN              77.14         NaN        5.7           8.57   \n",
              "5               1.92              63.46       11.54        7.7          13.46   \n",
              "6                NaN              80.00         NaN        NaN           8.89   \n",
              "7               4.69              57.81       15.62        9.4          14.06   \n",
              "8              22.50              55.00       42.50        7.5           5.00   \n",
              "9               7.14                NaN         NaN        NaN            NaN   \n",
              "\n",
              "  salt_100g  sodium_100g  nutrition-score-fr_100g nutrition-score-uk_100g  \n",
              "0       NaN          NaN                      NaN                     NaN  \n",
              "1   0.00000        0.000                     14.0                    14.0  \n",
              "2   0.63500        0.250                      0.0                     0.0  \n",
              "3   1.22428        0.482                     12.0                    12.0  \n",
              "4       NaN          NaN                      NaN                     NaN  \n",
              "5       NaN          NaN                      NaN                     NaN  \n",
              "6       NaN          NaN                      NaN                     NaN  \n",
              "7   0.13970        0.055                      7.0                     7.0  \n",
              "8       NaN          NaN                      NaN                     NaN  \n",
              "9       NaN          NaN                      NaN                     NaN  \n",
              "\n",
              "[10 rows x 34 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2f4c989f-8034-4701-9865-e182b9896f4c\">\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>code</th>\n",
              "      <th>url</th>\n",
              "      <th>creator</th>\n",
              "      <th>created_t</th>\n",
              "      <th>created_datetime</th>\n",
              "      <th>last_modified_t</th>\n",
              "      <th>last_modified_datetime</th>\n",
              "      <th>product_name</th>\n",
              "      <th>brands</th>\n",
              "      <th>brands_tags</th>\n",
              "      <th>...</th>\n",
              "      <th>fat_100g</th>\n",
              "      <th>saturated-fat_100g</th>\n",
              "      <th>carbohydrates_100g</th>\n",
              "      <th>sugars_100g</th>\n",
              "      <th>fiber_100g</th>\n",
              "      <th>proteins_100g</th>\n",
              "      <th>salt_100g</th>\n",
              "      <th>sodium_100g</th>\n",
              "      <th>nutrition-score-fr_100g</th>\n",
              "      <th>nutrition-score-uk_100g</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3087</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>openfoodfacts-contributors</td>\n",
              "      <td>1474103866</td>\n",
              "      <td>2016-09-17T09:17:46Z</td>\n",
              "      <td>1474103893</td>\n",
              "      <td>2016-09-17T09:18:13Z</td>\n",
              "      <td>Farine de blé noir</td>\n",
              "      <td>Ferme t'y R'nao</td>\n",
              "      <td>ferme-t-y-r-nao</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4530</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>Banana Chips Sweetened (Whole)</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>28.57</td>\n",
              "      <td>28.57</td>\n",
              "      <td>64.29</td>\n",
              "      <td>14.29</td>\n",
              "      <td>3.6</td>\n",
              "      <td>3.57</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>14.0</td>\n",
              "      <td>14.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4559</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>1489069957</td>\n",
              "      <td>2017-03-09T14:32:37Z</td>\n",
              "      <td>Peanuts</td>\n",
              "      <td>Torn &amp; Glasser</td>\n",
              "      <td>torn-glasser</td>\n",
              "      <td>...</td>\n",
              "      <td>17.86</td>\n",
              "      <td>0.00</td>\n",
              "      <td>60.71</td>\n",
              "      <td>17.86</td>\n",
              "      <td>7.1</td>\n",
              "      <td>17.86</td>\n",
              "      <td>0.63500</td>\n",
              "      <td>0.250</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>16087</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055731</td>\n",
              "      <td>2017-03-09T10:35:31Z</td>\n",
              "      <td>1489055731</td>\n",
              "      <td>2017-03-09T10:35:31Z</td>\n",
              "      <td>Organic Salted Nut Mix</td>\n",
              "      <td>Grizzlies</td>\n",
              "      <td>grizzlies</td>\n",
              "      <td>...</td>\n",
              "      <td>57.14</td>\n",
              "      <td>5.36</td>\n",
              "      <td>17.86</td>\n",
              "      <td>3.57</td>\n",
              "      <td>7.1</td>\n",
              "      <td>17.86</td>\n",
              "      <td>1.22428</td>\n",
              "      <td>0.482</td>\n",
              "      <td>12.0</td>\n",
              "      <td>12.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>16094</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055653</td>\n",
              "      <td>2017-03-09T10:34:13Z</td>\n",
              "      <td>1489055653</td>\n",
              "      <td>2017-03-09T10:34:13Z</td>\n",
              "      <td>Organic Polenta</td>\n",
              "      <td>Bob's Red Mill</td>\n",
              "      <td>bob-s-red-mill</td>\n",
              "      <td>...</td>\n",
              "      <td>1.43</td>\n",
              "      <td>NaN</td>\n",
              "      <td>77.14</td>\n",
              "      <td>NaN</td>\n",
              "      <td>5.7</td>\n",
              "      <td>8.57</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>16100</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055651</td>\n",
              "      <td>2017-03-09T10:34:11Z</td>\n",
              "      <td>1489055651</td>\n",
              "      <td>2017-03-09T10:34:11Z</td>\n",
              "      <td>Breadshop Honey Gone Nuts Granola</td>\n",
              "      <td>Unfi</td>\n",
              "      <td>unfi</td>\n",
              "      <td>...</td>\n",
              "      <td>18.27</td>\n",
              "      <td>1.92</td>\n",
              "      <td>63.46</td>\n",
              "      <td>11.54</td>\n",
              "      <td>7.7</td>\n",
              "      <td>13.46</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>16117</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055730</td>\n",
              "      <td>2017-03-09T10:35:30Z</td>\n",
              "      <td>1489055730</td>\n",
              "      <td>2017-03-09T10:35:30Z</td>\n",
              "      <td>Organic Long Grain White Rice</td>\n",
              "      <td>Lundberg</td>\n",
              "      <td>lundberg</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>80.00</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>8.89</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>16124</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055711</td>\n",
              "      <td>2017-03-09T10:35:11Z</td>\n",
              "      <td>1489055712</td>\n",
              "      <td>2017-03-09T10:35:12Z</td>\n",
              "      <td>Organic Muesli</td>\n",
              "      <td>Daddy's Muesli</td>\n",
              "      <td>daddy-s-muesli</td>\n",
              "      <td>...</td>\n",
              "      <td>18.75</td>\n",
              "      <td>4.69</td>\n",
              "      <td>57.81</td>\n",
              "      <td>15.62</td>\n",
              "      <td>9.4</td>\n",
              "      <td>14.06</td>\n",
              "      <td>0.13970</td>\n",
              "      <td>0.055</td>\n",
              "      <td>7.0</td>\n",
              "      <td>7.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>16193</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055651</td>\n",
              "      <td>2017-03-09T10:34:11Z</td>\n",
              "      <td>1489055651</td>\n",
              "      <td>2017-03-09T10:34:11Z</td>\n",
              "      <td>Organic Dark Chocolate Minis</td>\n",
              "      <td>Equal Exchange</td>\n",
              "      <td>equal-exchange</td>\n",
              "      <td>...</td>\n",
              "      <td>37.50</td>\n",
              "      <td>22.50</td>\n",
              "      <td>55.00</td>\n",
              "      <td>42.50</td>\n",
              "      <td>7.5</td>\n",
              "      <td>5.00</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>16513</td>\n",
              "      <td>http://world-en.openfoodfacts.org/product/0000...</td>\n",
              "      <td>usda-ndb-import</td>\n",
              "      <td>1489055654</td>\n",
              "      <td>2017-03-09T10:34:14Z</td>\n",
              "      <td>1489055654</td>\n",
              "      <td>2017-03-09T10:34:14Z</td>\n",
              "      <td>Organic Sunflower Oil</td>\n",
              "      <td>Napa Valley Naturals</td>\n",
              "      <td>napa-valley-naturals</td>\n",
              "      <td>...</td>\n",
              "      <td>100.00</td>\n",
              "      <td>7.14</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>10 rows × 34 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2f4c989f-8034-4701-9865-e182b9896f4c')\"\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-2f4c989f-8034-4701-9865-e182b9896f4c 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-2f4c989f-8034-4701-9865-e182b9896f4c');\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": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#check data types\n",
        "df_open_food.dtypes"
      ],
      "metadata": {
        "id": "dO9UlwW1shCb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "630e92a0-a1f4-4678-efbd-307815377751"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "code                                        object\n",
              "url                                         object\n",
              "creator                                     object\n",
              "created_t                                   object\n",
              "created_datetime                            object\n",
              "last_modified_t                             object\n",
              "last_modified_datetime                      object\n",
              "product_name                                object\n",
              "brands                                      object\n",
              "brands_tags                                 object\n",
              "countries                                   object\n",
              "countries_tags                              object\n",
              "countries_en                                object\n",
              "ingredients_text                            object\n",
              "serving_size                                object\n",
              "additives_n                                float64\n",
              "additives                                   object\n",
              "ingredients_from_palm_oil_n                float64\n",
              "ingredients_that_may_be_from_palm_oil_n    float64\n",
              "nutrition_grade_fr                          object\n",
              "states                                      object\n",
              "states_tags                                 object\n",
              "states_en                                   object\n",
              "energy_100g                                float64\n",
              "fat_100g                                   float64\n",
              "saturated-fat_100g                         float64\n",
              "carbohydrates_100g                         float64\n",
              "sugars_100g                                float64\n",
              "fiber_100g                                 float64\n",
              "proteins_100g                              float64\n",
              "salt_100g                                  float64\n",
              "sodium_100g                                float64\n",
              "nutrition-score-fr_100g                    float64\n",
              "nutrition-score-uk_100g                    float64\n",
              "dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "####Cleaning Country Column####\n",
        "Since we want to analyze where the food products are coming from, we first need to clean the country data. There are multiple columns with country information, including countries, country_tags, and countries_en. We ultimately decided to use countries_en, exploded the lists, and removed or edited duplicate names."
      ],
      "metadata": {
        "id": "nOxhQrV-Xr18"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Look into the 3 different countries columns and separate lists of countries\n",
        "score_country_df = df_open_food[['code', 'url', 'nutrition-score-fr_100g', 'countries_en', 'product_name']]\n",
        "\n",
        "score_country_df['country'] = score_country_df['countries_en'].str.split(\",|:\")\n",
        "score_country_df = score_country_df.explode('country')\n",
        "\n",
        "#clean up -- remove fr: or en: prefix, standardize duplicate countries\n",
        "score_country_df['country'] = score_country_df['country'].str.rsplit(\":\")\n",
        "score_country_df = score_country_df.explode('country')\n",
        "score_country_df = score_country_df[(score_country_df['country'] != 'fr') & \n",
        "                                    (score_country_df['country'] != 'en')]\n",
        "score_country_df['country'] = score_country_df['country'].replace({\n",
        "    'United-states-of-america': 'United States',\n",
        "    '日本': 'Japan', 'Other-日本':'Japan', 'Other-japon':'Japan',\n",
        "    'Republique-de-chine': 'China', '中华人民共和国' : 'China',\n",
        "    '香港':'Hong Kong',\n",
        "    'العراق':'Iraq','Other-العراق':'Iraq',\n",
        "    'ព្រះរាជាណាចក្រកម្ពុជា':'Cambodia',\n",
        "    'السعودية':'Saudi Arabia',\n",
        "    'المملكة-المتحدة':'United Kingdom',\n",
        "    'البحرين':'Bahrain',\n",
        "    'Франция' : 'France',\n",
        "    'سلطنة-عمان' : 'Oman',\n",
        "    'Other-turquie' : 'Turkey',\n",
        "    'Danemark' : 'Denmark',\n",
        "    'grande-bretagne' : 'United Kingdom', 'Reino-unido' : 'United Kingdom',\n",
        "    'Grande-bretagne': 'United Kingdom', 'Angleterre' : 'United Kingdom',\n",
        "    'भारत' : 'India', \n",
        "    'Marseille-5':'France', 'Belgien' : 'Belgium'\n",
        "    })\n",
        "\n",
        "#view unique countries\n",
        "uniq_country = pd.unique(score_country_df['country'])\n",
        "num_countries = uniq_country.size\n",
        "\n",
        "uniq_country \n",
        "#num_countries\n"
      ],
      "metadata": {
        "id": "mcAd0RSNtt5m",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1b79e9a7-65ba-4764-a2e0-a252c13c10ae"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-13-1a2c7a9ed15a>:4: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  score_country_df['country'] = score_country_df['countries_en'].str.split(\",|:\")\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['France', 'United States', 'United Kingdom', 'Canada',\n",
              "       'Switzerland', 'Spain', 'Germany', 'Slovenia', 'Bangladesh',\n",
              "       'Italy', 'Romania', 'Australia', 'Portugal', 'Belgium',\n",
              "       'Netherlands', 'Guadeloupe', 'Dominican Republic', 'Ireland',\n",
              "       'Réunion', 'Taiwan', 'Brazil', 'Mexico', 'Saudi Arabia',\n",
              "       'United Arab Emirates', 'Iraq', 'Kuwait', 'Moldova',\n",
              "       'New Caledonia', nan, 'Singapore', 'Finland', 'Sweden',\n",
              "       'French Polynesia', 'New Zealand', 'Thailand', 'Panama',\n",
              "       'Costa Rica', 'Scotland', 'Trinidad and Tobago', 'China',\n",
              "       'Hong Kong', 'Luxembourg', 'Lebanon', 'Denmark', 'Belgie',\n",
              "       'Nederland', 'Malaysia', 'Guyana', 'Serbia', 'Martinique',\n",
              "       'Barbados', 'Aruba', 'Norway', 'Haiti', 'El Salvador', 'Bahrain',\n",
              "       'Guatemala', 'Morocco', 'Oman', 'South Korea', 'Chile', 'Cuba',\n",
              "       'Japan', 'Israel', 'Philippines', 'Austria', 'Guinea', 'Jordan',\n",
              "       'Russia', 'Saint Pierre and Miquelon', 'Algeria', 'Quebec',\n",
              "       'Czech Republic', 'Mauritius', 'Indonesia', 'Armenia',\n",
              "       \"Côte d'Ivoire\", 'Poland', 'Greece', 'India', 'Ecuador', 'Turkey',\n",
              "       'Cambodia', 'Hungary', 'Argentina', 'European Union', 'Colombia',\n",
              "       'Qatar', 'French Guiana', 'Bulgaria', 'Burkina Faso', 'Peru',\n",
              "       'Tunisia', 'Puerto Rico', 'Nigeria', 'Honduras', 'Egypt', 'Monaco',\n",
              "       'Cyprus', 'Venezuela', 'South Africa', 'Mongolia', 'Myanmar',\n",
              "       'Vietnam', 'Suriname', 'Pakistan', 'Deutschland', 'Cameroon',\n",
              "       'Virgin Islands of the United States', 'Senegal', 'Slovakia',\n",
              "       'Lithuania', 'Togo', 'Iceland', 'Soviet Union', 'Global-market',\n",
              "       'Ukraine', 'World', 'Madagascar', 'Macau', 'Demonte-italie',\n",
              "       'Schweiz', 'Frankreich', 'Spanien', 'Albania', 'Malta', 'Croatia',\n",
              "       'Malawi', 'Mali', 'Gabon', 'Kazakhstan', 'Saint Martin',\n",
              "       'fresh-foods', 'dairies', 'Republic of the Congo', 'Mayotte',\n",
              "       'Alemania', 'Mauricio', 'Sint Maarten', 'Vanuatu', 'Latvia',\n",
              "       'Bourgogne-aube-nogent-sur-seine', 'Andorra', 'Puyricard',\n",
              "       'Ecosse', 'Wallis and Futuna', 'Frankrijk', 'Griekenland',\n",
              "       'Spanje', 'Estonia', 'Niger', 'Bosnia and Herzegovina',\n",
              "       'Republic of Macedonia', 'Belarus', 'Mauritania', 'Iran',\n",
              "       'Montenegro', 'Allemagne', 'Afghanistan', 'Azerbaijan', 'Chad',\n",
              "       'Sri Lanka', 'Georgia', 'Mozambique', 'Libya', 'Irland', 'Jersey',\n",
              "       'Burundi', 'Democratic Republic of the Congo', 'Ghana',\n",
              "       'Guinea-Bissau', 'Belgiqu', 'nl', 'Bolivia',\n",
              "       'Central African Republic', 'Angola', 'Szczecin', 'Namibia',\n",
              "       'Kenya', 'Tanzania', 'Rwanda', 'Brunei', 'Somalia', 'Syria',\n",
              "       'Yemen', 'State of Palestine', 'Palestinian territories', 'Ap',\n",
              "       'Isle of Man', 'Uruguay', 'Paraguay', 'Ethiopia', 'Cook Islands',\n",
              "       'Maldives', 'xx', 'Fiji'], dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **3** Exploratory Data Analysis"
      ],
      "metadata": {
        "id": "yZYjcTl5tbUk"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Countries vs Nutrition EDA\n",
        "Having cleaned the country column, we can now explore the origins of the food products in the dataset. First we list out the 10 most commonly occuring countries -- as expected, U.S. and France occur the most frequently."
      ],
      "metadata": {
        "id": "jg2EiHUuWAXs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Groupby, sorts, and head viewing of the countries and counts\n",
        "countries_df = score_country_df[['country','code']].groupby('country').count()\n",
        "countries_df = countries_df.sort_values('code', ascending = False)\n",
        "countries_df.rename(columns={'code': 'count'}, inplace=True)\n",
        "countries_df.head(5)"
      ],
      "metadata": {
        "id": "xWQDvM9xWJGe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "87c03400-fcde-4cbb-ae53-4ee16c205127"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                count\n",
              "country              \n",
              "United States  173709\n",
              "France         129322\n",
              "Switzerland     17204\n",
              "Germany          9403\n",
              "Spain            6061"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-af9cbe83-2c28-4f63-8e3c-f331249d4d19\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>count</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>country</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>United States</th>\n",
              "      <td>173709</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>France</th>\n",
              "      <td>129322</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Switzerland</th>\n",
              "      <td>17204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Germany</th>\n",
              "      <td>9403</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Spain</th>\n",
              "      <td>6061</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-af9cbe83-2c28-4f63-8e3c-f331249d4d19')\"\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-af9cbe83-2c28-4f63-8e3c-f331249d4d19 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-af9cbe83-2c28-4f63-8e3c-f331249d4d19');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Pie Chart of the Top 10 Countries represented"
      ],
      "metadata": {
        "id": "dOknlmdHb2eS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Create figure large enough so that the text is visible and doesn't overlap\n",
        "countries_df.head(10).plot(kind='pie', y='count', legend=False, figsize=(12, 12))\n",
        "plt.title(\"Proportion of Top 10 Countries in Dataset\", size = 15)"
      ],
      "metadata": {
        "id": "r4Bie21CbIBr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 718
        },
        "outputId": "5f409e15-055d-4f47-945f-cbbb5f65be5b"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Proportion of Top 10 Countries in Dataset')"
            ]
          },
          "metadata": {},
          "execution_count": 15
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x864 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Top 10 countries with best average nutrition score\n",
        "By averaging the nutrition score of all products for each country, we found the 10 countries in the dataset with the \"healthiest\" food."
      ],
      "metadata": {
        "id": "XwRBpURXXdBz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Mean and sort to find country with highest nutrition score\n",
        "nutrition_df = score_country_df[['country','code','nutrition-score-fr_100g']]\n",
        "nutrition_df = nutrition_df[nutrition_df['nutrition-score-fr_100g'].notna()]\n",
        "\n",
        "nutrition_df = nutrition_df.groupby('country')['nutrition-score-fr_100g'].mean().reset_index()\n",
        "nutrition_df = nutrition_df.sort_values('nutrition-score-fr_100g', ascending = False)\n",
        "nutrition_df.head(10)"
      ],
      "metadata": {
        "id": "L0TtURXpXm2u",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        },
        "outputId": "8d1f384b-42d7-415d-d23f-753f51ff6eed"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                country  nutrition-score-fr_100g\n",
              "91             Pakistan                33.000000\n",
              "12              Belarus                27.000000\n",
              "6               Armenia                22.000000\n",
              "132               Yemen                21.000000\n",
              "68                Libya                21.000000\n",
              "118  State of Palestine                21.000000\n",
              "63               Jordan                21.000000\n",
              "72           Madagascar                20.666667\n",
              "56            Indonesia                20.000000\n",
              "22             Cameroon                20.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-01903bfd-bf78-4781-b6d8-3131f45374b7\">\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>country</th>\n",
              "      <th>nutrition-score-fr_100g</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>91</th>\n",
              "      <td>Pakistan</td>\n",
              "      <td>33.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>Belarus</td>\n",
              "      <td>27.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Armenia</td>\n",
              "      <td>22.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>132</th>\n",
              "      <td>Yemen</td>\n",
              "      <td>21.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>68</th>\n",
              "      <td>Libya</td>\n",
              "      <td>21.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>118</th>\n",
              "      <td>State of Palestine</td>\n",
              "      <td>21.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>63</th>\n",
              "      <td>Jordan</td>\n",
              "      <td>21.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>72</th>\n",
              "      <td>Madagascar</td>\n",
              "      <td>20.666667</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>56</th>\n",
              "      <td>Indonesia</td>\n",
              "      <td>20.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>Cameroon</td>\n",
              "      <td>20.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-01903bfd-bf78-4781-b6d8-3131f45374b7')\"\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-01903bfd-bf78-4781-b6d8-3131f45374b7 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-01903bfd-bf78-4781-b6d8-3131f45374b7');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Looking at products and the ingredients within these products and brands"
      ],
      "metadata": {
        "id": "5Ekxp6U85BHC"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Products per Brand\n",
        "\n",
        "Looking at how many products each brand has on the market."
      ],
      "metadata": {
        "id": "0cTuFV3Vc_-U"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "First, let's take a look at how many products each brand has."
      ],
      "metadata": {
        "id": "6hl9dWaYd5gv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#looking at ingredients list of different products and brands\n",
        "prod_ingred_df = df_open_food[['code', 'url', 'nutrition-score-fr_100g', 'product_name', 'brands', 'brands_tags', 'ingredients_text', 'additives']]\n",
        "#look at the total number of brands represented\n",
        "brands_agg = prod_ingred_df.groupby(['brands_tags'])['brands_tags'].count()\n",
        "brands_agg.sort_values(ascending=False, inplace=True)\n",
        "brands_agg "
      ],
      "metadata": {
        "id": "Tjydfu3Wc8KA",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3b953547-3d17-4f8c-b028-e86c001c4fdf"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "brands_tags\n",
              "carrefour                        3847\n",
              "auchan                           3334\n",
              "u                                2637\n",
              "leader-price                     2228\n",
              "casino                           2178\n",
              "                                 ... \n",
              "jardins-de-carthage                 1\n",
              "jarfin-bio                          1\n",
              "jarritos                            1\n",
              "jarritos,the-jel-sert-company       1\n",
              "자연은,jayeon-eun                      1\n",
              "Name: brands_tags, Length: 56804, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "It seems like carrefour has the most brand tags with 3847. Let's look at some quick summary statistics on how many products each brand has."
      ],
      "metadata": {
        "id": "RWoHeY78eHCl"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#find the distribution of the data\n",
        "print(\"The mean is\", brands_agg.mean())\n",
        "print(\"The median is\", brands_agg.median())\n",
        "print(\"The variance is\", brands_agg.var())"
      ],
      "metadata": {
        "id": "g1jB3F90dJMU",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "62c64ed6-6c2d-4661-ad2f-911f6b6c82d5"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The mean is 5.755879867614957\n",
            "The median is 1.0\n",
            "The variance is 1609.941795083433\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Note the high variance in the summary statistics as well as the difference between the mean and the median. The difference between these two is most likely due to the high outlier brands that have a lot of products."
      ],
      "metadata": {
        "id": "SPLdEvW86ofC"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## The Most Common Ingredients\n",
        "Let's take a look at the ingredients list column and the most commonly used ingredients"
      ],
      "metadata": {
        "id": "hyKpwvCFeokZ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now, we'll look at the ingredients and which ones are most popular. We'll first convert the ingredients text into a string type for us to parse through."
      ],
      "metadata": {
        "id": "puDZD3rEePy5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#convert these into strings then lists then explode for ingredients\n",
        "prod_ingred_df['ingredients_text_str'] = prod_ingred_df['ingredients_text'].astype('str')\n",
        "prod_ingred_df['ingredients_text_str'] = prod_ingred_df['ingredients_text_str'].str.split(',')\n",
        "\n",
        "#separate the ingredients\n",
        "prod_ingred_exploded_df = prod_ingred_df.explode('ingredients_text_str')"
      ],
      "metadata": {
        "id": "bPeUbt36dK2E",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "29ffc73b-afd8-4c84-e82e-4538e29f6aa6"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-19-1f4897cdfdab>:2: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  prod_ingred_df['ingredients_text_str'] = prod_ingred_df['ingredients_text'].astype('str')\n",
            "<ipython-input-19-1f4897cdfdab>:3: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  prod_ingred_df['ingredients_text_str'] = prod_ingred_df['ingredients_text_str'].str.split(',')\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here we'll sort to find the top 10 most used ingredients and their counts."
      ],
      "metadata": {
        "id": "ApkDGXzt67tz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#find the top 10 most popular ingredients\n",
        "ingredients_plot = prod_ingred_exploded_df[prod_ingred_exploded_df['ingredients_text_str'] != 'nan']\n",
        "ingredients_plot = ingredients_plot[['brands_tags', 'ingredients_text_str']]\n",
        "ingredients_plot = ingredients_plot.groupby('ingredients_text_str').count().reset_index()\n",
        "ingredients_plot.sort_values(by='brands_tags', inplace=True, ascending=False)\n",
        "\n",
        "#show the top ten most used ingredients \n",
        "ingredients_plot.head(10)"
      ],
      "metadata": {
        "id": "rPXtD_v3dMU7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        },
        "outputId": "5622997d-2fdf-4efe-f87e-3eb978e35e46"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       ingredients_text_str  brands_tags\n",
              "342000                 salt        81933\n",
              "373331                sugar        51861\n",
              "397032                water        47959\n",
              "347998                  sel        39508\n",
              "370926                sucre        25850\n",
              "138991          citric acid        24229\n",
              "182644                  eau        23265\n",
              "280403               niacin        18915\n",
              "176266             dextrose        17817\n",
              "336485           riboflavin        17011"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-594fac16-7344-46aa-b92a-cd08fd7c9bbf\">\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>ingredients_text_str</th>\n",
              "      <th>brands_tags</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>342000</th>\n",
              "      <td>salt</td>\n",
              "      <td>81933</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>373331</th>\n",
              "      <td>sugar</td>\n",
              "      <td>51861</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>397032</th>\n",
              "      <td>water</td>\n",
              "      <td>47959</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>347998</th>\n",
              "      <td>sel</td>\n",
              "      <td>39508</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>370926</th>\n",
              "      <td>sucre</td>\n",
              "      <td>25850</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>138991</th>\n",
              "      <td>citric acid</td>\n",
              "      <td>24229</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>182644</th>\n",
              "      <td>eau</td>\n",
              "      <td>23265</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>280403</th>\n",
              "      <td>niacin</td>\n",
              "      <td>18915</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>176266</th>\n",
              "      <td>dextrose</td>\n",
              "      <td>17817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>336485</th>\n",
              "      <td>riboflavin</td>\n",
              "      <td>17011</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-594fac16-7344-46aa-b92a-cd08fd7c9bbf')\"\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-594fac16-7344-46aa-b92a-cd08fd7c9bbf 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-594fac16-7344-46aa-b92a-cd08fd7c9bbf');\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": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Ingredient Word Cloud\n",
        "\n",
        "Let's visualize this in a fun wordcloud using nltk!"
      ],
      "metadata": {
        "id": "WnVdrh3ZeruD"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Importing all the necessary components of our word cloud."
      ],
      "metadata": {
        "id": "UQ2Zbs5W7H5u"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import nltk\n",
        "nltk.download('punkt')\n",
        "nltk.__version__"
      ],
      "metadata": {
        "id": "ra-RvygRdNwa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        },
        "outputId": "4f163a48-5669-4fe6-8c21-8d04a99bb594"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'3.7'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from nltk.corpus import stopwords\n",
        "\n",
        "#Do stopword analysis\n",
        "nltk.download('stopwords')\n",
        "stopwords = set(stopwords.words('english'))"
      ],
      "metadata": {
        "id": "OchN6Wgle2ER",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a0f03582-3d65-4d0e-fa4b-1382a89440b8"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Creation of our tokenize content function."
      ],
      "metadata": {
        "id": "KXdUrJ_87Euw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def tokenize_content(content):\n",
        "  #tokenize text\n",
        "  words = nltk.word_tokenize(content)\n",
        "  #remove tokens that contain alphabetic characters \n",
        "  words = [word for word in words if word.isalpha()]\n",
        "  #only alphabetic characters and convert them to lower case \n",
        "  words = [word.lower() for word in words if word.isalpha()]\n",
        "  #remove stopwords\n",
        "  words = [word for word in words if word not in stopwords]\n",
        "  return words"
      ],
      "metadata": {
        "id": "5cgk3RYhdSFf"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#convert ingredient names to string\n",
        "prod_ingred_df['ingredients_text_whole_str'] = prod_ingred_df['ingredients_text'].astype('str')\n",
        "ingredients_names = prod_ingred_df[prod_ingred_df['ingredients_text_whole_str'] != 'nan']\n",
        "ingredients_list = ingredients_names['ingredients_text_whole_str']\n",
        "\n",
        "top_tokens_list = []\n",
        "for brand in ingredients_list:\n",
        "  top_tokens_list.append(tokenize_content(brand))\n",
        "\n",
        "#flatten the list\n",
        "top_tokens = []\n",
        "for sublist in top_tokens_list:\n",
        "  for item in sublist:\n",
        "    top_tokens.append(item)"
      ],
      "metadata": {
        "id": "Qm7f8YhpdUh7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "23a53fef-df86-4963-dbe2-d0f3a03741da"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-24-d2749e6e7125>:2: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  prod_ingred_df['ingredients_text_whole_str'] = prod_ingred_df['ingredients_text'].astype('str')\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Below we have the 20 most common ingredients used in products based on our tokenization."
      ],
      "metadata": {
        "id": "pu7L90Xe7MAJ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Find 20 most common words\n",
        "import collections as col\n",
        "top_most_common = col.Counter(top_tokens).most_common(20)\n",
        "top_most_common"
      ],
      "metadata": {
        "id": "eZepJYqjddff",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "87026d42-5e80-40a6-bff2-345247d25f5e"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('de', 409600),\n",
              " ('salt', 133267),\n",
              " ('sugar', 108374),\n",
              " ('oil', 96889),\n",
              " ('acid', 94887),\n",
              " ('water', 91747),\n",
              " ('flour', 80104),\n",
              " ('organic', 77116),\n",
              " ('corn', 73220),\n",
              " ('sodium', 72502),\n",
              " ('sel', 69209),\n",
              " ('milk', 68506),\n",
              " ('natural', 65858),\n",
              " ('wheat', 56744),\n",
              " ('sucre', 55847),\n",
              " ('syrup', 49353),\n",
              " ('eau', 45979),\n",
              " ('flavor', 45469),\n",
              " ('et', 41758),\n",
              " ('powder', 40998)]"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's plot our wordcloud!"
      ],
      "metadata": {
        "id": "6vQG187C7Q9f"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#make a word cloud for top tokens \n",
        "from wordcloud import WordCloud\n",
        "#make the string \n",
        "text = ' '.join(top_tokens)\n",
        "\n",
        "#feed it through the wordcloud function\n",
        "wordcloud = WordCloud(width = 800, height = 800, min_font_size = 10, background_color = 'white').generate(text)\n",
        "\n",
        "#show the figure\n",
        "plt.figure(figsize=(10, 10))\n",
        "plt.imshow(wordcloud, interpolation='bilinear')\n",
        "plt.axis(\"off\")\n",
        "plt.tight_layout(pad = 0)\n",
        "plt.title(\"Word Cloud of Top Ingredient Words\")\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "Btjiglqndet8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 767
        },
        "outputId": "00f8b472-a6c1-4c0a-b8ff-8a414e2a2223"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## EDA for Nutrition"
      ],
      "metadata": {
        "id": "OM6YWlugsIph"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We have access to data about nutrition information, including energy, macronutrients (carbs, fat, protein), and other nutrients as well."
      ],
      "metadata": {
        "id": "VArnhUIG-Dfw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's rename some columns to make them easier to deal with"
      ],
      "metadata": {
        "id": "koF3sC1B-PUT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Drop '100g' from each column name\n",
        "df_open_food = df_open_food.rename(axis=1, mapper={\"energy_100g\": \"energy\", \n",
        "                                                   \"fat_100g\" : \"fat\", \n",
        "                                                   \"saturated-fat_100g\" : \"saturated_fat\", \n",
        "                                                   \"carbohydrates_100g\" : \"carbohydrates\", \n",
        "                                                   \"sugars_100g\": \"sugars\", \n",
        "                                                   \"fiber_100g\": \"fiber\", \n",
        "                                                   \"proteins_100g\" : \"protein\", \n",
        "                                                   \"salt_100g\": \"salt\",\n",
        "                                                   \"sodium_100g\": \"sodium\"\n",
        "                                                   })\n",
        "print(df_open_food.dtypes)"
      ],
      "metadata": {
        "id": "q7uLWiJysNgt",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fa9de79b-e976-4ff4-e49d-39ee0f35a1da"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "code                                        object\n",
            "url                                         object\n",
            "creator                                     object\n",
            "created_t                                   object\n",
            "created_datetime                            object\n",
            "last_modified_t                             object\n",
            "last_modified_datetime                      object\n",
            "product_name                                object\n",
            "brands                                      object\n",
            "brands_tags                                 object\n",
            "countries                                   object\n",
            "countries_tags                              object\n",
            "countries_en                                object\n",
            "ingredients_text                            object\n",
            "serving_size                                object\n",
            "additives_n                                float64\n",
            "additives                                   object\n",
            "ingredients_from_palm_oil_n                float64\n",
            "ingredients_that_may_be_from_palm_oil_n    float64\n",
            "nutrition_grade_fr                          object\n",
            "states                                      object\n",
            "states_tags                                 object\n",
            "states_en                                   object\n",
            "energy                                     float64\n",
            "fat                                        float64\n",
            "saturated_fat                              float64\n",
            "carbohydrates                              float64\n",
            "sugars                                     float64\n",
            "fiber                                      float64\n",
            "protein                                    float64\n",
            "salt                                       float64\n",
            "sodium                                     float64\n",
            "nutrition-score-fr_100g                    float64\n",
            "nutrition-score-uk_100g                    float64\n",
            "dtype: object\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We want to make sure all our data is valid, so we can drop any instance with a negative or overly large nutrition value. Ultimately, this does catch a few outliers that don't make practical sense. We also drop any instance with a NaN value."
      ],
      "metadata": {
        "id": "U15Q_zwl-Ubk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Removing anything above 1_000_000 and below 0 for either entry as they are not valid data (there exist negative values and and values > 1e6)\n",
        "df_nutrients = df_open_food[[\"energy\", \"fat\", \"saturated_fat\", \"carbohydrates\", \"sugars\", \"fiber\", \"protein\", \"salt\", \"sodium\"]]\n",
        "df_nutrients = df_nutrients[(df_nutrients['energy'] >= 0) & (df_nutrients['energy'] <= 1_000_000) & \n",
        "                            (df_nutrients['fat'] >= 0) & (df_nutrients['fat'] <= 1_000_000) &\n",
        "                            (df_nutrients['saturated_fat'] >= 0) & (df_nutrients['saturated_fat'] <= 1_000_000) & \n",
        "                            (df_nutrients['carbohydrates'] >= 0) & (df_nutrients['carbohydrates'] <= 1_000_000) &\n",
        "                            (df_nutrients['sugars'] >= 0) & (df_nutrients['sugars'] <= 1_000_000) &\n",
        "                            (df_nutrients['fiber'] >= 0) & (df_nutrients['fiber'] <= 1_000_000) &\n",
        "                            (df_nutrients['protein'] >= 0) & (df_nutrients['protein'] <= 1_000_000) & \n",
        "                            (df_nutrients['salt'] >= 0) & (df_nutrients['salt'] <= 1_000_000) &\n",
        "                            (df_nutrients['sodium'] >= 0) & (df_nutrients['sodium'] <= 1_000_000) \n",
        "                            ]\n",
        "df_nutrients = df_nutrients.dropna(how=\"any\")\n",
        "df_nutrients.size"
      ],
      "metadata": {
        "id": "d1CB3J6uv--h",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a4954af5-2209-4ade-9ae9-3a74021b55c7"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1780641"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Creating a boxplot of all nutrients (excluding energy since it's not a nutrient and is larger than the other elements). We will remove outliers for now as they make the boxplot difficult to see"
      ],
      "metadata": {
        "id": "ljf3wXc8zUxj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_nutrients_plot = df_nutrients.drop(columns=\"energy\")\n",
        "\n",
        "#Remove outliers and set labels to be rotated for improved readability\n",
        "ax_nutrients = sns.boxplot(data=df_nutrients_plot, showfliers=False)\n",
        "ax_nutrients.set_xticklabels(ax_nutrients.get_xticklabels(), rotation=40, ha=\"right\")\n",
        "plt.title(\"Boxplots of All Nutrients\" , size = 10 )\n",
        "plt.ylabel(\"grams per 100g\" , size = 10 )\n",
        "plt.xlabel(\"Nutrient\" , size = 10 )\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "yYNcnTPhzUiP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "outputId": "7955a922-029b-42d2-9332-006dc840a680"
      },
      "execution_count": 29,
      "outputs": [
        {
          "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": [
        "Overall we see a healthy distribution with most foods having a median close to zero (which makes sense since a lot of foods don't have that ingredient, like spicy foods and sugar). In the case of fiber, salt, and sodium, the nutrients are overall quite low, whereas for sugars, carbohydrates, and fats, there are foods with high values."
      ],
      "metadata": {
        "id": "BwTa9Syy-_lZ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Note that the macronutrients (fat, carbohydrates, and protein) are 3 of the features that have the biggest distribution. So, let's analyze their distribution further below."
      ],
      "metadata": {
        "id": "z8FANKA3ubFE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Creating macronutrients_df\n",
        "df_macronutrients = df_nutrients[[\"fat\", \"carbohydrates\", \"protein\"]]\n",
        "#sns.boxplot(data=df_macronutrients)\n",
        "#Excluding outliers to get a better picture \n",
        "sns.boxplot(data=df_macronutrients, showfliers=False)\n",
        "plt.title(\"Macronutrient Content\")\n",
        "plt.ylabel(\"grams per 100g\" , size = 10 )\n",
        "plt.xlabel(\"Macronutrient\" , size = 10 )"
      ],
      "metadata": {
        "id": "4gITjn47uIU1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "cd7a1899-eaaa-47af-dbff-c272c418d343"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Macronutrient')"
            ]
          },
          "metadata": {},
          "execution_count": 30
        },
        {
          "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": [
        "There is a lot of variability in the carbohydrates, and a medium amount in the fat, but less in the protein. Also note the high median for carbs. This indicates that foods tend to have more carbs than the other 2 macronutrients, but also that the variability is high among foods for carb levels"
      ],
      "metadata": {
        "id": "jhFm7Emm_hNq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can also analyze correlations but creating a correlational heatmap of the relations between the various nutrition metrics"
      ],
      "metadata": {
        "id": "rN1XHdmD2qaZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "corr = df_nutrients.corr()\n",
        "\n",
        "#Coolwarm is a good colorstyle to visualize with\n",
        "corr.style.background_gradient(cmap ='coolwarm')\n"
      ],
      "metadata": {
        "id": "D-AyVMIv2mlM",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 332
        },
        "outputId": "32cb998d-8bfd-454e-c268-cfd057d8667f"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<pandas.io.formats.style.Styler at 0x7f64ee5674c0>"
            ],
            "text/html": [
              "<style type=\"text/css\">\n",
              "#T_80eac_row0_col0, #T_80eac_row1_col1, #T_80eac_row2_col2, #T_80eac_row3_col3, #T_80eac_row4_col4, #T_80eac_row5_col5, #T_80eac_row6_col6, #T_80eac_row7_col7, #T_80eac_row7_col8, #T_80eac_row8_col7, #T_80eac_row8_col8 {\n",
              "  background-color: #b40426;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row0_col1 {\n",
              "  background-color: #f18f71;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row0_col2, #T_80eac_row2_col0 {\n",
              "  background-color: #ebd3c6;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row0_col3 {\n",
              "  background-color: #f0cdbb;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row0_col4 {\n",
              "  background-color: #d4dbe6;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row0_col5 {\n",
              "  background-color: #81a4fb;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row0_col6 {\n",
              "  background-color: #cdd9ec;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row0_col7, #T_80eac_row0_col8, #T_80eac_row2_col7, #T_80eac_row2_col8 {\n",
              "  background-color: #3f53c6;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row1_col0 {\n",
              "  background-color: #f29274;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row1_col2 {\n",
              "  background-color: #f7aa8c;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row1_col3 {\n",
              "  background-color: #485fd1;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row1_col4 {\n",
              "  background-color: #7b9ff9;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row1_col5 {\n",
              "  background-color: #5e7de7;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row1_col6 {\n",
              "  background-color: #c4d5f3;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row1_col7, #T_80eac_row1_col8, #T_80eac_row2_col5, #T_80eac_row5_col2 {\n",
              "  background-color: #4055c8;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row2_col1 {\n",
              "  background-color: #f7a889;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row2_col3 {\n",
              "  background-color: #516ddb;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row2_col4, #T_80eac_row6_col0 {\n",
              "  background-color: #a3c2fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row2_col6 {\n",
              "  background-color: #adc9fd;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row3_col0 {\n",
              "  background-color: #ead5c9;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row3_col1, #T_80eac_row3_col2, #T_80eac_row4_col5, #T_80eac_row4_col6, #T_80eac_row4_col7, #T_80eac_row4_col8, #T_80eac_row6_col3, #T_80eac_row6_col4, #T_80eac_row7_col0, #T_80eac_row7_col2, #T_80eac_row8_col0, #T_80eac_row8_col2 {\n",
              "  background-color: #3b4cc0;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row3_col4 {\n",
              "  background-color: #f7af91;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row3_col5, #T_80eac_row5_col4 {\n",
              "  background-color: #779af7;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row3_col6 {\n",
              "  background-color: #6180e9;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row3_col7, #T_80eac_row3_col8, #T_80eac_row7_col5, #T_80eac_row8_col5 {\n",
              "  background-color: #3e51c5;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row4_col0, #T_80eac_row5_col6 {\n",
              "  background-color: #abc8fd;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row4_col1 {\n",
              "  background-color: #4961d2;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row4_col2 {\n",
              "  background-color: #6b8df0;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row4_col3 {\n",
              "  background-color: #f6bda2;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row5_col0 {\n",
              "  background-color: #80a3fa;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row5_col1 {\n",
              "  background-color: #6788ee;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row5_col3 {\n",
              "  background-color: #8db0fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row5_col7, #T_80eac_row5_col8 {\n",
              "  background-color: #4257c9;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row6_col1 {\n",
              "  background-color: #9ebeff;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_80eac_row6_col2 {\n",
              "  background-color: #7699f6;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row6_col5 {\n",
              "  background-color: #7597f6;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row6_col7, #T_80eac_row6_col8, #T_80eac_row7_col1, #T_80eac_row8_col1 {\n",
              "  background-color: #465ecf;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row7_col3, #T_80eac_row8_col3 {\n",
              "  background-color: #506bda;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row7_col4, #T_80eac_row8_col4 {\n",
              "  background-color: #7396f5;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_80eac_row7_col6, #T_80eac_row8_col6 {\n",
              "  background-color: #7ea1fa;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "</style>\n",
              "<table id=\"T_80eac_\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th class=\"blank level0\" >&nbsp;</th>\n",
              "      <th class=\"col_heading level0 col0\" >energy</th>\n",
              "      <th class=\"col_heading level0 col1\" >fat</th>\n",
              "      <th class=\"col_heading level0 col2\" >saturated_fat</th>\n",
              "      <th class=\"col_heading level0 col3\" >carbohydrates</th>\n",
              "      <th class=\"col_heading level0 col4\" >sugars</th>\n",
              "      <th class=\"col_heading level0 col5\" >fiber</th>\n",
              "      <th class=\"col_heading level0 col6\" >protein</th>\n",
              "      <th class=\"col_heading level0 col7\" >salt</th>\n",
              "      <th class=\"col_heading level0 col8\" >sodium</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row0\" class=\"row_heading level0 row0\" >energy</th>\n",
              "      <td id=\"T_80eac_row0_col0\" class=\"data row0 col0\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row0_col1\" class=\"data row0 col1\" >0.765860</td>\n",
              "      <td id=\"T_80eac_row0_col2\" class=\"data row0 col2\" >0.555783</td>\n",
              "      <td id=\"T_80eac_row0_col3\" class=\"data row0 col3\" >0.547135</td>\n",
              "      <td id=\"T_80eac_row0_col4\" class=\"data row0 col4\" >0.332232</td>\n",
              "      <td id=\"T_80eac_row0_col5\" class=\"data row0 col5\" >0.204843</td>\n",
              "      <td id=\"T_80eac_row0_col6\" class=\"data row0 col6\" >0.305759</td>\n",
              "      <td id=\"T_80eac_row0_col7\" class=\"data row0 col7\" >-0.010552</td>\n",
              "      <td id=\"T_80eac_row0_col8\" class=\"data row0 col8\" >-0.010552</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row1\" class=\"row_heading level0 row1\" >fat</th>\n",
              "      <td id=\"T_80eac_row1_col0\" class=\"data row1 col0\" >0.765860</td>\n",
              "      <td id=\"T_80eac_row1_col1\" class=\"data row1 col1\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row1_col2\" class=\"data row1 col2\" >0.700154</td>\n",
              "      <td id=\"T_80eac_row1_col3\" class=\"data row1 col3\" >-0.047538</td>\n",
              "      <td id=\"T_80eac_row1_col4\" class=\"data row1 col4\" >0.002212</td>\n",
              "      <td id=\"T_80eac_row1_col5\" class=\"data row1 col5\" >0.102577</td>\n",
              "      <td id=\"T_80eac_row1_col6\" class=\"data row1 col6\" >0.263769</td>\n",
              "      <td id=\"T_80eac_row1_col7\" class=\"data row1 col7\" >-0.005765</td>\n",
              "      <td id=\"T_80eac_row1_col8\" class=\"data row1 col8\" >-0.005765</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row2\" class=\"row_heading level0 row2\" >saturated_fat</th>\n",
              "      <td id=\"T_80eac_row2_col0\" class=\"data row2 col0\" >0.555783</td>\n",
              "      <td id=\"T_80eac_row2_col1\" class=\"data row2 col1\" >0.700154</td>\n",
              "      <td id=\"T_80eac_row2_col2\" class=\"data row2 col2\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row2_col3\" class=\"data row2 col3\" >-0.010566</td>\n",
              "      <td id=\"T_80eac_row2_col4\" class=\"data row2 col4\" >0.143723</td>\n",
              "      <td id=\"T_80eac_row2_col5\" class=\"data row2 col5\" >0.009250</td>\n",
              "      <td id=\"T_80eac_row2_col6\" class=\"data row2 col6\" >0.177162</td>\n",
              "      <td id=\"T_80eac_row2_col7\" class=\"data row2 col7\" >-0.008495</td>\n",
              "      <td id=\"T_80eac_row2_col8\" class=\"data row2 col8\" >-0.008495</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row3\" class=\"row_heading level0 row3\" >carbohydrates</th>\n",
              "      <td id=\"T_80eac_row3_col0\" class=\"data row3 col0\" >0.547135</td>\n",
              "      <td id=\"T_80eac_row3_col1\" class=\"data row3 col1\" >-0.047538</td>\n",
              "      <td id=\"T_80eac_row3_col2\" class=\"data row3 col2\" >-0.010566</td>\n",
              "      <td id=\"T_80eac_row3_col3\" class=\"data row3 col3\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row3_col4\" class=\"data row3 col4\" >0.617922</td>\n",
              "      <td id=\"T_80eac_row3_col5\" class=\"data row3 col5\" >0.179102</td>\n",
              "      <td id=\"T_80eac_row3_col6\" class=\"data row3 col6\" >-0.095576</td>\n",
              "      <td id=\"T_80eac_row3_col7\" class=\"data row3 col7\" >-0.014517</td>\n",
              "      <td id=\"T_80eac_row3_col8\" class=\"data row3 col8\" >-0.014517</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row4\" class=\"row_heading level0 row4\" >sugars</th>\n",
              "      <td id=\"T_80eac_row4_col0\" class=\"data row4 col0\" >0.332232</td>\n",
              "      <td id=\"T_80eac_row4_col1\" class=\"data row4 col1\" >0.002212</td>\n",
              "      <td id=\"T_80eac_row4_col2\" class=\"data row4 col2\" >0.143723</td>\n",
              "      <td id=\"T_80eac_row4_col3\" class=\"data row4 col3\" >0.617922</td>\n",
              "      <td id=\"T_80eac_row4_col4\" class=\"data row4 col4\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row4_col5\" class=\"data row4 col5\" >-0.013310</td>\n",
              "      <td id=\"T_80eac_row4_col6\" class=\"data row4 col6\" >-0.248632</td>\n",
              "      <td id=\"T_80eac_row4_col7\" class=\"data row4 col7\" >-0.027672</td>\n",
              "      <td id=\"T_80eac_row4_col8\" class=\"data row4 col8\" >-0.027672</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row5\" class=\"row_heading level0 row5\" >fiber</th>\n",
              "      <td id=\"T_80eac_row5_col0\" class=\"data row5 col0\" >0.204843</td>\n",
              "      <td id=\"T_80eac_row5_col1\" class=\"data row5 col1\" >0.102577</td>\n",
              "      <td id=\"T_80eac_row5_col2\" class=\"data row5 col2\" >0.009250</td>\n",
              "      <td id=\"T_80eac_row5_col3\" class=\"data row5 col3\" >0.179102</td>\n",
              "      <td id=\"T_80eac_row5_col4\" class=\"data row5 col4\" >-0.013310</td>\n",
              "      <td id=\"T_80eac_row5_col5\" class=\"data row5 col5\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row5_col6\" class=\"data row5 col6\" >0.171301</td>\n",
              "      <td id=\"T_80eac_row5_col7\" class=\"data row5 col7\" >-0.000985</td>\n",
              "      <td id=\"T_80eac_row5_col8\" class=\"data row5 col8\" >-0.000985</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row6\" class=\"row_heading level0 row6\" >protein</th>\n",
              "      <td id=\"T_80eac_row6_col0\" class=\"data row6 col0\" >0.305759</td>\n",
              "      <td id=\"T_80eac_row6_col1\" class=\"data row6 col1\" >0.263769</td>\n",
              "      <td id=\"T_80eac_row6_col2\" class=\"data row6 col2\" >0.177162</td>\n",
              "      <td id=\"T_80eac_row6_col3\" class=\"data row6 col3\" >-0.095576</td>\n",
              "      <td id=\"T_80eac_row6_col4\" class=\"data row6 col4\" >-0.248632</td>\n",
              "      <td id=\"T_80eac_row6_col5\" class=\"data row6 col5\" >0.171301</td>\n",
              "      <td id=\"T_80eac_row6_col6\" class=\"data row6 col6\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row6_col7\" class=\"data row6 col7\" >0.013709</td>\n",
              "      <td id=\"T_80eac_row6_col8\" class=\"data row6 col8\" >0.013708</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row7\" class=\"row_heading level0 row7\" >salt</th>\n",
              "      <td id=\"T_80eac_row7_col0\" class=\"data row7 col0\" >-0.010552</td>\n",
              "      <td id=\"T_80eac_row7_col1\" class=\"data row7 col1\" >-0.005765</td>\n",
              "      <td id=\"T_80eac_row7_col2\" class=\"data row7 col2\" >-0.008495</td>\n",
              "      <td id=\"T_80eac_row7_col3\" class=\"data row7 col3\" >-0.014517</td>\n",
              "      <td id=\"T_80eac_row7_col4\" class=\"data row7 col4\" >-0.027672</td>\n",
              "      <td id=\"T_80eac_row7_col5\" class=\"data row7 col5\" >-0.000985</td>\n",
              "      <td id=\"T_80eac_row7_col6\" class=\"data row7 col6\" >0.013709</td>\n",
              "      <td id=\"T_80eac_row7_col7\" class=\"data row7 col7\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row7_col8\" class=\"data row7 col8\" >1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_80eac_level0_row8\" class=\"row_heading level0 row8\" >sodium</th>\n",
              "      <td id=\"T_80eac_row8_col0\" class=\"data row8 col0\" >-0.010552</td>\n",
              "      <td id=\"T_80eac_row8_col1\" class=\"data row8 col1\" >-0.005765</td>\n",
              "      <td id=\"T_80eac_row8_col2\" class=\"data row8 col2\" >-0.008495</td>\n",
              "      <td id=\"T_80eac_row8_col3\" class=\"data row8 col3\" >-0.014517</td>\n",
              "      <td id=\"T_80eac_row8_col4\" class=\"data row8 col4\" >-0.027672</td>\n",
              "      <td id=\"T_80eac_row8_col5\" class=\"data row8 col5\" >-0.000985</td>\n",
              "      <td id=\"T_80eac_row8_col6\" class=\"data row8 col6\" >0.013708</td>\n",
              "      <td id=\"T_80eac_row8_col7\" class=\"data row8 col7\" >1.000000</td>\n",
              "      <td id=\"T_80eac_row8_col8\" class=\"data row8 col8\" >1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We see that salt and sodium are essentially the same item! This is important when we do machine learning analysis and look at feature selection. Also, energy seems to be correlated with the other macronutrients. This also makes sense, since carbs and protein are 4 calories/gram, and protein is 9 cals/gram, showing a direct relationship between these nutrients and energy. "
      ],
      "metadata": {
        "id": "1d6zLoW1_6FF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Again, notice how several variables are highly correlated with each other, such as sodium and salt (which are essentially the same thing), fat and energy, saturated_fat and fat, and sugars and carbohydrates."
      ],
      "metadata": {
        "id": "Ix_dlRibnpJ4"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **4** Machine Learning to Determine Healthy Products"
      ],
      "metadata": {
        "id": "MODXpgse0KPB"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In this section, we are going to use a variety of supervised machine learning methods to regress on various features to determine their nutrition score. Then we are going to classify on several features to determine whether we can predict the \"healthiness\" of a product."
      ],
      "metadata": {
        "id": "4bLrmUPonZdP"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Imports"
      ],
      "metadata": {
        "id": "DQAsPi6MMQSc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.linear_model import LinearRegression\n",
        "from sklearn.linear_model import Ridge\n",
        "from sklearn.linear_model import RidgeCV\n",
        "from sklearn.linear_model import Lasso\n",
        "from sklearn.linear_model import LassoCV\n",
        "from sklearn.linear_model import ElasticNet\n",
        "from sklearn.linear_model import ElasticNetCV\n",
        "from sklearn.decomposition import PCA\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "#Import random forest, neural networks\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.neural_network import MLPClassifier\n",
        "\n",
        "#Import pytorch neural network classes\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "\n",
        "#Import helpers\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.metrics import mean_squared_error\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.model_selection import KFold\n",
        "from sklearn.model_selection import cross_val_score\n"
      ],
      "metadata": {
        "id": "a4fzljIlMSIT"
      },
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We need to convert our response variable into a categorical variable that ML algorithms can predict the nutrition score. We also need to make sure that we remove any features that we may not need. We mostly want to keep information about the nutritional value of these food products. We are using the nutrition-score-fr_100g as the nutrition score index.\n",
        "\n",
        "We will first clean in a very similar way done in the nutrition EDA"
      ],
      "metadata": {
        "id": "QpkFFmBT0sW1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Removing anything above 1_000_000 and below 0 for either entry as they are not valid data (there exist negative values and and values > 1e6)\n",
        "df_open_food_nan = df_open_food[['additives_n', 'ingredients_from_palm_oil_n', 'ingredients_that_may_be_from_palm_oil_n', \n",
        "                                'energy', 'fat', 'saturated_fat', 'carbohydrates', 'sugars', 'fiber', 'protein', 'salt', 'sodium','nutrition-score-fr_100g']]\n",
        "df_open_food_nan = df_open_food_nan[(df_open_food_nan['energy'] >= 0) & (df_open_food_nan['energy'] <= 1_000_000) & \n",
        "                            (df_open_food_nan['fat'] >= 0) & (df_open_food_nan['fat'] <= 1_000_000) &\n",
        "                            (df_open_food_nan['saturated_fat'] >= 0) & (df_open_food_nan['saturated_fat'] <= 1_000_000) & \n",
        "                            (df_open_food_nan['carbohydrates'] >= 0) & (df_open_food_nan['carbohydrates'] <= 1_000_000) &\n",
        "                            (df_open_food_nan['sugars'] >= 0) & (df_open_food_nan['sugars'] <= 1_000_000) &\n",
        "                            (df_open_food_nan['fiber'] >= 0) & (df_open_food_nan['fiber'] <= 1_000_000) &\n",
        "                            (df_open_food_nan['protein'] >= 0) & (df_open_food_nan['protein'] <= 1_000_000) & \n",
        "                            (df_open_food_nan['salt'] >= 0) & (df_open_food_nan['salt'] <= 1_000_000) &\n",
        "                            (df_open_food_nan['sodium'] >= 0) & (df_open_food_nan['sodium'] <= 1_000_000) \n",
        "                            ]\n",
        "\n",
        "#also drop any Nan values\n",
        "df_open_food_nan = df_open_food_nan.dropna(how=\"any\")\n",
        "\n",
        "# select helpful features for our machine learning algorithm\n",
        "df_open_food_ml = df_open_food_nan[['additives_n', 'ingredients_from_palm_oil_n', 'ingredients_that_may_be_from_palm_oil_n', \n",
        "                                'energy', 'fat', 'saturated_fat', 'carbohydrates', 'sugars', 'fiber', 'protein', 'salt', 'sodium', 'nutrition-score-fr_100g']]\n",
        "\n",
        "#see the size of the dataset\n",
        "len(df_open_food_nan)"
      ],
      "metadata": {
        "id": "4AHSTm-R0SCR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4f7904cd-f227-4a6e-97c7-2ea11627a2bf"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "190770"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now we need to split the nutrition score where we categorize foods that have a score higher or equal to the median is \"healthy\" (1), and anything below is \"unhealthy\" (0). This is because we believe that health_score is a good indicator of a nutrition score.   \n",
        "\n",
        "This variable will be used for classification analyses, whereas the score itself can be used for regression and real-number analyses."
      ],
      "metadata": {
        "id": "ksVDfEvU7SvB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#finding the median score of the nutrition score\n",
        "median_score = df_open_food_ml[['nutrition-score-fr_100g']].median()\n",
        "\n",
        "#creating the categorical response variable\n",
        "df_open_food_ml['health_score'] = np.where(df_open_food_ml[['nutrition-score-fr_100g']] >= median_score, 'healthy', 'unhealthy')\n",
        "df_open_food_ml['nutrition-score-fr_100g'].describe()"
      ],
      "metadata": {
        "id": "KYjKJLEz61qv",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6932f702-303a-49d4-f870-861f3e527a98"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "additives_n                                145117\n",
            "ingredients_from_palm_oil_n                145117\n",
            "ingredients_that_may_be_from_palm_oil_n    145117\n",
            "energy                                     145117\n",
            "fat                                        145117\n",
            "saturated_fat                              145117\n",
            "carbohydrates                              145117\n",
            "sugars                                     145117\n",
            "fiber                                      145117\n",
            "protein                                    145117\n",
            "salt                                       145117\n",
            "sodium                                     145117\n",
            "nutrition-score-fr_100g                    145117\n",
            "health_score                               145117\n",
            "dtype: int64\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "count    190770.000000\n",
              "mean          8.672910\n",
              "std           9.027616\n",
              "min         -15.000000\n",
              "25%           1.000000\n",
              "50%           9.000000\n",
              "75%          16.000000\n",
              "max          40.000000\n",
              "Name: nutrition-score-fr_100g, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Separating Features and Response"
      ],
      "metadata": {
        "id": "nlH8V_fdIg3E"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#stores features dataframe into variable called \"features\"\n",
        "features = df_open_food_ml.loc[:, ~df_open_food_ml.columns.isin(['nutrition-score-fr_100g', 'health_score'])]\n",
        "\n"
      ],
      "metadata": {
        "id": "aVeeQ1UEJdw_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#check correlation between features\n",
        "corr = features.corr()\n",
        "corr.style.background_gradient(cmap ='coolwarm')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 446
        },
        "id": "-Od9gjhYveR_",
        "outputId": "9b7bb578-24b8-4f76-846b-c1c811efdb93"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<pandas.io.formats.style.Styler at 0x7ff51b1486d0>"
            ],
            "text/html": [
              "<style type=\"text/css\">\n",
              "#T_44634_row0_col0, #T_44634_row1_col1, #T_44634_row2_col2, #T_44634_row3_col3, #T_44634_row4_col4, #T_44634_row5_col5, #T_44634_row6_col6, #T_44634_row7_col7, #T_44634_row8_col8, #T_44634_row9_col9, #T_44634_row10_col10, #T_44634_row10_col11, #T_44634_row11_col10, #T_44634_row11_col11 {\n",
              "  background-color: #b40426;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row0_col1 {\n",
              "  background-color: #6384eb;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row0_col2, #T_44634_row0_col7 {\n",
              "  background-color: #a7c5fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row0_col3, #T_44634_row0_col10, #T_44634_row0_col11, #T_44634_row2_col3, #T_44634_row6_col4, #T_44634_row8_col5 {\n",
              "  background-color: #4358cb;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row0_col4, #T_44634_row0_col5, #T_44634_row0_col8, #T_44634_row7_col9, #T_44634_row7_col10, #T_44634_row7_col11, #T_44634_row8_col0, #T_44634_row9_col1, #T_44634_row9_col2, #T_44634_row9_col6, #T_44634_row9_col7, #T_44634_row10_col3, #T_44634_row11_col3 {\n",
              "  background-color: #3b4cc0;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row0_col6 {\n",
              "  background-color: #88abfd;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row0_col9 {\n",
              "  background-color: #5f7fe8;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col0 {\n",
              "  background-color: #7da0f9;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col2 {\n",
              "  background-color: #81a4fb;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col3, #T_44634_row2_col6, #T_44634_row5_col8 {\n",
              "  background-color: #5b7ae5;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col4 {\n",
              "  background-color: #6282ea;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col5, #T_44634_row1_col8, #T_44634_row5_col1, #T_44634_row10_col0, #T_44634_row11_col0 {\n",
              "  background-color: #5a78e4;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col6 {\n",
              "  background-color: #6f92f3;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col7 {\n",
              "  background-color: #8db0fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row1_col9, #T_44634_row2_col1, #T_44634_row8_col7 {\n",
              "  background-color: #779af7;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row1_col10, #T_44634_row1_col11, #T_44634_row2_col10, #T_44634_row2_col11, #T_44634_row4_col0, #T_44634_row8_col10, #T_44634_row8_col11 {\n",
              "  background-color: #4257c9;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row2_col0 {\n",
              "  background-color: #b3cdfb;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row2_col4 {\n",
              "  background-color: #5673e0;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row2_col5 {\n",
              "  background-color: #4a63d3;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row2_col7, #T_44634_row10_col9, #T_44634_row11_col9 {\n",
              "  background-color: #80a3fa;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row2_col8, #T_44634_row4_col1, #T_44634_row10_col4, #T_44634_row11_col4 {\n",
              "  background-color: #506bda;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row2_col9 {\n",
              "  background-color: #6e90f2;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row3_col0 {\n",
              "  background-color: #5e7de7;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row3_col1 {\n",
              "  background-color: #5d7ce6;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row3_col2, #T_44634_row6_col2 {\n",
              "  background-color: #4e68d8;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row3_col4 {\n",
              "  background-color: #f08b6e;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row3_col5 {\n",
              "  background-color: #ecd3c5;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row3_col6 {\n",
              "  background-color: #f0cdbb;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row3_col7 {\n",
              "  background-color: #d4dbe6;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row3_col8 {\n",
              "  background-color: #9bbcff;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row3_col9 {\n",
              "  background-color: #cedaeb;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row3_col10, #T_44634_row3_col11, #T_44634_row5_col10, #T_44634_row5_col11, #T_44634_row8_col1, #T_44634_row8_col2 {\n",
              "  background-color: #3f53c6;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row4_col2 {\n",
              "  background-color: #4c66d6;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row4_col3 {\n",
              "  background-color: #f29274;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row4_col5 {\n",
              "  background-color: #f7aa8c;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row4_col6 {\n",
              "  background-color: #485fd1;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row4_col7 {\n",
              "  background-color: #7b9ff9;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row4_col8 {\n",
              "  background-color: #7a9df8;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row4_col9 {\n",
              "  background-color: #c5d6f2;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row4_col10, #T_44634_row4_col11 {\n",
              "  background-color: #4055c8;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row5_col0 {\n",
              "  background-color: #5470de;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row5_col2, #T_44634_row5_col6, #T_44634_row10_col6, #T_44634_row11_col6 {\n",
              "  background-color: #536edd;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row5_col3 {\n",
              "  background-color: #ebd3c6;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row5_col4 {\n",
              "  background-color: #f6a586;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row5_col7, #T_44634_row9_col3 {\n",
              "  background-color: #a3c2fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row5_col9, #T_44634_row8_col9 {\n",
              "  background-color: #aec9fc;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row6_col0 {\n",
              "  background-color: #8badfd;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row6_col1, #T_44634_row10_col8, #T_44634_row11_col8 {\n",
              "  background-color: #5875e1;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row6_col3 {\n",
              "  background-color: #ead5c9;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row6_col5, #T_44634_row9_col0 {\n",
              "  background-color: #3c4ec2;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row6_col7 {\n",
              "  background-color: #f7ad90;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row6_col8 {\n",
              "  background-color: #93b5fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row6_col9 {\n",
              "  background-color: #6180e9;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row6_col10, #T_44634_row6_col11, #T_44634_row10_col1, #T_44634_row11_col1 {\n",
              "  background-color: #3e51c5;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row7_col0 {\n",
              "  background-color: #89acfd;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row7_col1, #T_44634_row7_col4 {\n",
              "  background-color: #516ddb;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row7_col2 {\n",
              "  background-color: #4b64d5;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row7_col3 {\n",
              "  background-color: #abc8fd;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row7_col5 {\n",
              "  background-color: #6b8df0;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row7_col6 {\n",
              "  background-color: #f7bca1;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row7_col8 {\n",
              "  background-color: #5572df;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row8_col3 {\n",
              "  background-color: #82a6fb;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row8_col4 {\n",
              "  background-color: #7396f5;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row8_col6, #T_44634_row9_col8 {\n",
              "  background-color: #90b2fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row9_col4 {\n",
              "  background-color: #a6c4fe;\n",
              "  color: #000000;\n",
              "}\n",
              "#T_44634_row9_col5 {\n",
              "  background-color: #799cf8;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row9_col10, #T_44634_row9_col11, #T_44634_row10_col2, #T_44634_row11_col2 {\n",
              "  background-color: #465ecf;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row10_col5, #T_44634_row11_col5 {\n",
              "  background-color: #3d50c3;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "#T_44634_row10_col7, #T_44634_row11_col7 {\n",
              "  background-color: #7597f6;\n",
              "  color: #f1f1f1;\n",
              "}\n",
              "</style>\n",
              "<table id=\"T_44634_\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th class=\"blank level0\" >&nbsp;</th>\n",
              "      <th class=\"col_heading level0 col0\" >additives_n</th>\n",
              "      <th class=\"col_heading level0 col1\" >ingredients_from_palm_oil_n</th>\n",
              "      <th class=\"col_heading level0 col2\" >ingredients_that_may_be_from_palm_oil_n</th>\n",
              "      <th class=\"col_heading level0 col3\" >energy</th>\n",
              "      <th class=\"col_heading level0 col4\" >fat</th>\n",
              "      <th class=\"col_heading level0 col5\" >saturated_fat</th>\n",
              "      <th class=\"col_heading level0 col6\" >carbohydrates</th>\n",
              "      <th class=\"col_heading level0 col7\" >sugars</th>\n",
              "      <th class=\"col_heading level0 col8\" >fiber</th>\n",
              "      <th class=\"col_heading level0 col9\" >protein</th>\n",
              "      <th class=\"col_heading level0 col10\" >salt</th>\n",
              "      <th class=\"col_heading level0 col11\" >sodium</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row0\" class=\"row_heading level0 row0\" >additives_n</th>\n",
              "      <td id=\"T_44634_row0_col0\" class=\"data row0 col0\" >1.000000</td>\n",
              "      <td id=\"T_44634_row0_col1\" class=\"data row0 col1\" >0.117145</td>\n",
              "      <td id=\"T_44634_row0_col2\" class=\"data row0 col2\" >0.291698</td>\n",
              "      <td id=\"T_44634_row0_col3\" class=\"data row0 col3\" >0.018059</td>\n",
              "      <td id=\"T_44634_row0_col4\" class=\"data row0 col4\" >-0.083061</td>\n",
              "      <td id=\"T_44634_row0_col5\" class=\"data row0 col5\" >-0.017723</td>\n",
              "      <td id=\"T_44634_row0_col6\" class=\"data row0 col6\" >0.159443</td>\n",
              "      <td id=\"T_44634_row0_col7\" class=\"data row0 col7\" >0.155856</td>\n",
              "      <td id=\"T_44634_row0_col8\" class=\"data row0 col8\" >-0.110501</td>\n",
              "      <td id=\"T_44634_row0_col9\" class=\"data row0 col9\" >-0.102294</td>\n",
              "      <td id=\"T_44634_row0_col10\" class=\"data row0 col10\" >0.002580</td>\n",
              "      <td id=\"T_44634_row0_col11\" class=\"data row0 col11\" >0.002580</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row1\" class=\"row_heading level0 row1\" >ingredients_from_palm_oil_n</th>\n",
              "      <td id=\"T_44634_row1_col0\" class=\"data row1 col0\" >0.117145</td>\n",
              "      <td id=\"T_44634_row1_col1\" class=\"data row1 col1\" >1.000000</td>\n",
              "      <td id=\"T_44634_row1_col2\" class=\"data row1 col2\" >0.177724</td>\n",
              "      <td id=\"T_44634_row1_col3\" class=\"data row1 col3\" >0.097651</td>\n",
              "      <td id=\"T_44634_row1_col4\" class=\"data row1 col4\" >0.056455</td>\n",
              "      <td id=\"T_44634_row1_col5\" class=\"data row1 col5\" >0.087469</td>\n",
              "      <td id=\"T_44634_row1_col6\" class=\"data row1 col6\" >0.082595</td>\n",
              "      <td id=\"T_44634_row1_col7\" class=\"data row1 col7\" >0.061974</td>\n",
              "      <td id=\"T_44634_row1_col8\" class=\"data row1 col8\" >0.003197</td>\n",
              "      <td id=\"T_44634_row1_col9\" class=\"data row1 col9\" >-0.015889</td>\n",
              "      <td id=\"T_44634_row1_col10\" class=\"data row1 col10\" >-0.002056</td>\n",
              "      <td id=\"T_44634_row1_col11\" class=\"data row1 col11\" >-0.002056</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row2\" class=\"row_heading level0 row2\" >ingredients_that_may_be_from_palm_oil_n</th>\n",
              "      <td id=\"T_44634_row2_col0\" class=\"data row2 col0\" >0.291698</td>\n",
              "      <td id=\"T_44634_row2_col1\" class=\"data row2 col1\" >0.177724</td>\n",
              "      <td id=\"T_44634_row2_col2\" class=\"data row2 col2\" >1.000000</td>\n",
              "      <td id=\"T_44634_row2_col3\" class=\"data row2 col3\" >0.019535</td>\n",
              "      <td id=\"T_44634_row2_col4\" class=\"data row2 col4\" >0.015635</td>\n",
              "      <td id=\"T_44634_row2_col5\" class=\"data row2 col5\" >0.035809</td>\n",
              "      <td id=\"T_44634_row2_col6\" class=\"data row2 col6\" >0.018420</td>\n",
              "      <td id=\"T_44634_row2_col7\" class=\"data row2 col7\" >0.012664</td>\n",
              "      <td id=\"T_44634_row2_col8\" class=\"data row2 col8\" >-0.028205</td>\n",
              "      <td id=\"T_44634_row2_col9\" class=\"data row2 col9\" >-0.048269</td>\n",
              "      <td id=\"T_44634_row2_col10\" class=\"data row2 col10\" >-0.003513</td>\n",
              "      <td id=\"T_44634_row2_col11\" class=\"data row2 col11\" >-0.003513</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row3\" class=\"row_heading level0 row3\" >energy</th>\n",
              "      <td id=\"T_44634_row3_col0\" class=\"data row3 col0\" >0.018059</td>\n",
              "      <td id=\"T_44634_row3_col1\" class=\"data row3 col1\" >0.097651</td>\n",
              "      <td id=\"T_44634_row3_col2\" class=\"data row3 col2\" >0.019535</td>\n",
              "      <td id=\"T_44634_row3_col3\" class=\"data row3 col3\" >1.000000</td>\n",
              "      <td id=\"T_44634_row3_col4\" class=\"data row3 col4\" >0.767079</td>\n",
              "      <td id=\"T_44634_row3_col5\" class=\"data row3 col5\" >0.555481</td>\n",
              "      <td id=\"T_44634_row3_col6\" class=\"data row3 col6\" >0.547232</td>\n",
              "      <td id=\"T_44634_row3_col7\" class=\"data row3 col7\" >0.330476</td>\n",
              "      <td id=\"T_44634_row3_col8\" class=\"data row3 col8\" >0.212487</td>\n",
              "      <td id=\"T_44634_row3_col9\" class=\"data row3 col9\" >0.307014</td>\n",
              "      <td id=\"T_44634_row3_col10\" class=\"data row3 col10\" >-0.009974</td>\n",
              "      <td id=\"T_44634_row3_col11\" class=\"data row3 col11\" >-0.009974</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row4\" class=\"row_heading level0 row4\" >fat</th>\n",
              "      <td id=\"T_44634_row4_col0\" class=\"data row4 col0\" >-0.083061</td>\n",
              "      <td id=\"T_44634_row4_col1\" class=\"data row4 col1\" >0.056455</td>\n",
              "      <td id=\"T_44634_row4_col2\" class=\"data row4 col2\" >0.015635</td>\n",
              "      <td id=\"T_44634_row4_col3\" class=\"data row4 col3\" >0.767079</td>\n",
              "      <td id=\"T_44634_row4_col4\" class=\"data row4 col4\" >1.000000</td>\n",
              "      <td id=\"T_44634_row4_col5\" class=\"data row4 col5\" >0.699610</td>\n",
              "      <td id=\"T_44634_row4_col6\" class=\"data row4 col6\" >-0.049427</td>\n",
              "      <td id=\"T_44634_row4_col7\" class=\"data row4 col7\" >-0.001881</td>\n",
              "      <td id=\"T_44634_row4_col8\" class=\"data row4 col8\" >0.109145</td>\n",
              "      <td id=\"T_44634_row4_col9\" class=\"data row4 col9\" >0.267144</td>\n",
              "      <td id=\"T_44634_row4_col10\" class=\"data row4 col10\" >-0.005800</td>\n",
              "      <td id=\"T_44634_row4_col11\" class=\"data row4 col11\" >-0.005800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row5\" class=\"row_heading level0 row5\" >saturated_fat</th>\n",
              "      <td id=\"T_44634_row5_col0\" class=\"data row5 col0\" >-0.017723</td>\n",
              "      <td id=\"T_44634_row5_col1\" class=\"data row5 col1\" >0.087469</td>\n",
              "      <td id=\"T_44634_row5_col2\" class=\"data row5 col2\" >0.035809</td>\n",
              "      <td id=\"T_44634_row5_col3\" class=\"data row5 col3\" >0.555481</td>\n",
              "      <td id=\"T_44634_row5_col4\" class=\"data row5 col4\" >0.699610</td>\n",
              "      <td id=\"T_44634_row5_col5\" class=\"data row5 col5\" >1.000000</td>\n",
              "      <td id=\"T_44634_row5_col6\" class=\"data row5 col6\" >-0.013148</td>\n",
              "      <td id=\"T_44634_row5_col7\" class=\"data row5 col7\" >0.140358</td>\n",
              "      <td id=\"T_44634_row5_col8\" class=\"data row5 col8\" >0.010702</td>\n",
              "      <td id=\"T_44634_row5_col9\" class=\"data row5 col9\" >0.178800</td>\n",
              "      <td id=\"T_44634_row5_col10\" class=\"data row5 col10\" >-0.008931</td>\n",
              "      <td id=\"T_44634_row5_col11\" class=\"data row5 col11\" >-0.008931</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row6\" class=\"row_heading level0 row6\" >carbohydrates</th>\n",
              "      <td id=\"T_44634_row6_col0\" class=\"data row6 col0\" >0.159443</td>\n",
              "      <td id=\"T_44634_row6_col1\" class=\"data row6 col1\" >0.082595</td>\n",
              "      <td id=\"T_44634_row6_col2\" class=\"data row6 col2\" >0.018420</td>\n",
              "      <td id=\"T_44634_row6_col3\" class=\"data row6 col3\" >0.547232</td>\n",
              "      <td id=\"T_44634_row6_col4\" class=\"data row6 col4\" >-0.049427</td>\n",
              "      <td id=\"T_44634_row6_col5\" class=\"data row6 col5\" >-0.013148</td>\n",
              "      <td id=\"T_44634_row6_col6\" class=\"data row6 col6\" >1.000000</td>\n",
              "      <td id=\"T_44634_row6_col7\" class=\"data row6 col7\" >0.619447</td>\n",
              "      <td id=\"T_44634_row6_col8\" class=\"data row6 col8\" >0.185316</td>\n",
              "      <td id=\"T_44634_row6_col9\" class=\"data row6 col9\" >-0.100950</td>\n",
              "      <td id=\"T_44634_row6_col10\" class=\"data row6 col10\" >-0.014654</td>\n",
              "      <td id=\"T_44634_row6_col11\" class=\"data row6 col11\" >-0.014654</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row7\" class=\"row_heading level0 row7\" >sugars</th>\n",
              "      <td id=\"T_44634_row7_col0\" class=\"data row7 col0\" >0.155856</td>\n",
              "      <td id=\"T_44634_row7_col1\" class=\"data row7 col1\" >0.061974</td>\n",
              "      <td id=\"T_44634_row7_col2\" class=\"data row7 col2\" >0.012664</td>\n",
              "      <td id=\"T_44634_row7_col3\" class=\"data row7 col3\" >0.330476</td>\n",
              "      <td id=\"T_44634_row7_col4\" class=\"data row7 col4\" >-0.001881</td>\n",
              "      <td id=\"T_44634_row7_col5\" class=\"data row7 col5\" >0.140358</td>\n",
              "      <td id=\"T_44634_row7_col6\" class=\"data row7 col6\" >0.619447</td>\n",
              "      <td id=\"T_44634_row7_col7\" class=\"data row7 col7\" >1.000000</td>\n",
              "      <td id=\"T_44634_row7_col8\" class=\"data row7 col8\" >-0.014625</td>\n",
              "      <td id=\"T_44634_row7_col9\" class=\"data row7 col9\" >-0.252968</td>\n",
              "      <td id=\"T_44634_row7_col10\" class=\"data row7 col10\" >-0.027699</td>\n",
              "      <td id=\"T_44634_row7_col11\" class=\"data row7 col11\" >-0.027699</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row8\" class=\"row_heading level0 row8\" >fiber</th>\n",
              "      <td id=\"T_44634_row8_col0\" class=\"data row8 col0\" >-0.110501</td>\n",
              "      <td id=\"T_44634_row8_col1\" class=\"data row8 col1\" >0.003197</td>\n",
              "      <td id=\"T_44634_row8_col2\" class=\"data row8 col2\" >-0.028205</td>\n",
              "      <td id=\"T_44634_row8_col3\" class=\"data row8 col3\" >0.212487</td>\n",
              "      <td id=\"T_44634_row8_col4\" class=\"data row8 col4\" >0.109145</td>\n",
              "      <td id=\"T_44634_row8_col5\" class=\"data row8 col5\" >0.010702</td>\n",
              "      <td id=\"T_44634_row8_col6\" class=\"data row8 col6\" >0.185316</td>\n",
              "      <td id=\"T_44634_row8_col7\" class=\"data row8 col7\" >-0.014625</td>\n",
              "      <td id=\"T_44634_row8_col8\" class=\"data row8 col8\" >1.000000</td>\n",
              "      <td id=\"T_44634_row8_col9\" class=\"data row8 col9\" >0.177822</td>\n",
              "      <td id=\"T_44634_row8_col10\" class=\"data row8 col10\" >-0.003291</td>\n",
              "      <td id=\"T_44634_row8_col11\" class=\"data row8 col11\" >-0.003291</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row9\" class=\"row_heading level0 row9\" >protein</th>\n",
              "      <td id=\"T_44634_row9_col0\" class=\"data row9 col0\" >-0.102294</td>\n",
              "      <td id=\"T_44634_row9_col1\" class=\"data row9 col1\" >-0.015889</td>\n",
              "      <td id=\"T_44634_row9_col2\" class=\"data row9 col2\" >-0.048269</td>\n",
              "      <td id=\"T_44634_row9_col3\" class=\"data row9 col3\" >0.307014</td>\n",
              "      <td id=\"T_44634_row9_col4\" class=\"data row9 col4\" >0.267144</td>\n",
              "      <td id=\"T_44634_row9_col5\" class=\"data row9 col5\" >0.178800</td>\n",
              "      <td id=\"T_44634_row9_col6\" class=\"data row9 col6\" >-0.100950</td>\n",
              "      <td id=\"T_44634_row9_col7\" class=\"data row9 col7\" >-0.252968</td>\n",
              "      <td id=\"T_44634_row9_col8\" class=\"data row9 col8\" >0.177822</td>\n",
              "      <td id=\"T_44634_row9_col9\" class=\"data row9 col9\" >1.000000</td>\n",
              "      <td id=\"T_44634_row9_col10\" class=\"data row9 col10\" >0.013052</td>\n",
              "      <td id=\"T_44634_row9_col11\" class=\"data row9 col11\" >0.013052</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row10\" class=\"row_heading level0 row10\" >salt</th>\n",
              "      <td id=\"T_44634_row10_col0\" class=\"data row10 col0\" >0.002580</td>\n",
              "      <td id=\"T_44634_row10_col1\" class=\"data row10 col1\" >-0.002056</td>\n",
              "      <td id=\"T_44634_row10_col2\" class=\"data row10 col2\" >-0.003513</td>\n",
              "      <td id=\"T_44634_row10_col3\" class=\"data row10 col3\" >-0.009974</td>\n",
              "      <td id=\"T_44634_row10_col4\" class=\"data row10 col4\" >-0.005800</td>\n",
              "      <td id=\"T_44634_row10_col5\" class=\"data row10 col5\" >-0.008931</td>\n",
              "      <td id=\"T_44634_row10_col6\" class=\"data row10 col6\" >-0.014654</td>\n",
              "      <td id=\"T_44634_row10_col7\" class=\"data row10 col7\" >-0.027699</td>\n",
              "      <td id=\"T_44634_row10_col8\" class=\"data row10 col8\" >-0.003291</td>\n",
              "      <td id=\"T_44634_row10_col9\" class=\"data row10 col9\" >0.013052</td>\n",
              "      <td id=\"T_44634_row10_col10\" class=\"data row10 col10\" >1.000000</td>\n",
              "      <td id=\"T_44634_row10_col11\" class=\"data row10 col11\" >1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_44634_level0_row11\" class=\"row_heading level0 row11\" >sodium</th>\n",
              "      <td id=\"T_44634_row11_col0\" class=\"data row11 col0\" >0.002580</td>\n",
              "      <td id=\"T_44634_row11_col1\" class=\"data row11 col1\" >-0.002056</td>\n",
              "      <td id=\"T_44634_row11_col2\" class=\"data row11 col2\" >-0.003513</td>\n",
              "      <td id=\"T_44634_row11_col3\" class=\"data row11 col3\" >-0.009974</td>\n",
              "      <td id=\"T_44634_row11_col4\" class=\"data row11 col4\" >-0.005800</td>\n",
              "      <td id=\"T_44634_row11_col5\" class=\"data row11 col5\" >-0.008931</td>\n",
              "      <td id=\"T_44634_row11_col6\" class=\"data row11 col6\" >-0.014654</td>\n",
              "      <td id=\"T_44634_row11_col7\" class=\"data row11 col7\" >-0.027699</td>\n",
              "      <td id=\"T_44634_row11_col8\" class=\"data row11 col8\" >-0.003291</td>\n",
              "      <td id=\"T_44634_row11_col9\" class=\"data row11 col9\" >0.013052</td>\n",
              "      <td id=\"T_44634_row11_col10\" class=\"data row11 col10\" >1.000000</td>\n",
              "      <td id=\"T_44634_row11_col11\" class=\"data row11 col11\" >1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Through this heat map we are able to see that there are several features that are highly correlated with each other. For example, sodium and salt are nearly identical, and fat and energy are highly correlated. As a result, we will be removing them from the features used."
      ],
      "metadata": {
        "id": "joK5uWMnwr7G"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#dropped features with high correlations\n",
        "features = features.drop(columns = ['sodium', 'ingredients_from_palm_oil_n', 'energy'])\n",
        "\n",
        "#store the classification target variable\n",
        "nutrition_score = df_open_food_ml['health_score']\n",
        "\n",
        "#store the regression target variable \n",
        "nutrition_regression = df_open_food_ml['nutrition-score-fr_100g']"
      ],
      "metadata": {
        "id": "p3oN-pwqw7-7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Splitting into Training and Test Set For Regression"
      ],
      "metadata": {
        "id": "6sYBKuWGK8qi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Assign appropriate value to seed and conduct 80/20 train-test split with random_state = seed \n",
        "seed = 42\n",
        "X = features\n",
        "y = nutrition_regression\n",
        "\n",
        "x_train, x_test, y_train, y_test = train_test_split(X, y, shuffle=True,\n",
        "                                   test_size=0.20, random_state=seed)"
      ],
      "metadata": {
        "id": "GVWasYxKLI4o"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **4.1** Regression Models to Predict Nutrition Scores"
      ],
      "metadata": {
        "id": "bc0QQa-KLRkj"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In this section we are going to use several models to regress on our nutrition score (nutrition-score-fr_100g) to see which model is the most effective."
      ],
      "metadata": {
        "id": "_6SFBJumm9Wy"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Unregularized OLS Regression"
      ],
      "metadata": {
        "id": "XJYh7PeQLuZc"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Firstly, we are going to create a linear regression without regularization to see how well we are able to predict a health score based on a simple OLS without regularization.  \n",
        "\n",
        "This is a good baseline to check for weights of features, scores, and mean squared error."
      ],
      "metadata": {
        "id": "PzaTI11sXYWg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# initialize the model and fit it on the training set\n",
        "reg = LinearRegression()\n",
        "reg.fit(x_train, y_train)\n",
        "\n",
        "# predict on the test set \n",
        "y_pred = reg.predict(x_test)\n",
        "\n",
        "# calculate the R-squared score\n",
        "score = reg.score(x_train, y_train)\n",
        "test_score = reg.score(x_test, y_test)\n",
        "print(\"R^2 score for TRAIN: \" + str(score))\n",
        "print(\"R^2 score for TEST: \" + str(test_score))\n",
        "\n",
        "# calculate the MSE\n",
        "print(\"MSE for TEST: \" + str(mean_squared_error(y_pred, y_test)))\n",
        "\n",
        "# estimated coefficients\n",
        "print(\"Coefficients determined: \" + str(reg.coef_))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "s3nfDi8ZL2xT",
        "outputId": "404f0c2b-fc51-4e4b-f96b-0468f69a7328"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "R^2 score for TRAIN: 0.6649442567890541\n",
            "R^2 score for TEST: 0.6742535216195189\n",
            "MSE for TEST: 26.44689636308995\n",
            "Coefficients determined: [ 0.50719621 -0.61790369  0.18149993  0.4489247   0.01402693  0.19159877\n",
            " -0.21162959  0.10941375  0.03342987]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The $R^2$ scores are decent, and the MSE is quite low. This means we are off by avout 5, on average. The coefficients are alsot mostly significant, and sending them to zero is not desired. So we predict using Ridge will do better than Lasso"
      ],
      "metadata": {
        "id": "hbtRRIN0CS_U"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Ridge Regression"
      ],
      "metadata": {
        "id": "2PcQeFCnNef_"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We use the `RidgeCV` class in to perform $L_2$ Regularized Linear Regression. We initialize a Ridge regression model named `reg_ridge`, with alpha set to 10.0 through cross validation.\n",
        "\n",
        "We saved our $R^2$ score as a variable called `ridge_score`."
      ],
      "metadata": {
        "id": "atplt0jBWasx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# initialize the Ridge model and fit it on the training set using cross-validation\n",
        "reg_ridge = RidgeCV()\n",
        "reg_ridge.fit(x_train, y_train)\n",
        "\n",
        "# predict on the test set \n",
        "y_pred = reg_ridge.predict(x_test)\n",
        "\n",
        "# calculate the R-squared score\n",
        "ridge_score = reg_ridge.score(x_train, y_train)\n",
        "ridge_test_score = reg_ridge.score(x_test, y_test)\n",
        "print(\"R^2 score for TRAIN: \" + str(ridge_score))\n",
        "print(\"R^2 score for TEST: \" + str(ridge_test_score))\n",
        "\n",
        "# calculate the MSE\n",
        "print(\"MSE for TEST: \" + str(mean_squared_error(y_pred, y_test)))\n",
        "\n",
        "# estimated coefficients\n",
        "print(\"Coefficients determined: \" + str(reg_ridge.coef_))\n",
        "print(\"Alpha determined: \" + str(reg_ridge.alpha_))"
      ],
      "metadata": {
        "id": "MHMcQ1ZQNgZi",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d5772878-79ed-47e3-974d-9fc357171290"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "R^2 score for TRAIN: 0.6649442564594649\n",
            "R^2 score for TEST: 0.6742538378393168\n",
            "MSE for TEST: 26.446870689651323\n",
            "Coefficients determined: [ 0.50717065 -0.61726094  0.18149987  0.44892287  0.014027    0.19159926\n",
            " -0.21162976  0.10941451  0.03342991]\n",
            "Alpha determined: 10.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Overall: Similar (but not equal!) performance. Also, note the similar $R^2$ scores between train and test in both cases, which is evidence that there is no significant under/overfitting"
      ],
      "metadata": {
        "id": "Jq9wQyEYDQAl"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Lasso Regression "
      ],
      "metadata": {
        "id": "6DPPYNdrNg9-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We use the `LassoCV` class in to perform $L_1$ Regularized Linear Regression. We initialize a Ridge regression model named `reg_lassocv`. Through cross validation, our alpha was automatically set to approx. 4.22.\n",
        "\n",
        "We saved our $R^2$ score as a variable called `lassocv_score`."
      ],
      "metadata": {
        "id": "-2rKQlSzW6hI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# initialize the model with alpha = 5 and fit it on the training set\n",
        "reg_lassocv = LassoCV()\n",
        "reg_lassocv.fit(x_train, y_train)\n",
        "\n",
        "# predict on the test set \n",
        "y_pred = reg_lassocv.predict(x_test)\n",
        "\n",
        "# calculate the R-squared score\n",
        "lassocv_score = reg_lassocv.score(x_train, y_train)\n",
        "lassocv_test_score = reg_lassocv.score(x_test, y_test)\n",
        "print(\"R^2 score for TRAIN: \" + str(lassocv_score))\n",
        "print(\"R^2 score for TEST: \" + str(lassocv_test_score))\n",
        "\n",
        "# calculate the MSE\n",
        "print(\"MSE for TEST: \" + str(mean_squared_error(y_pred, y_test)))\n",
        "\n",
        "# estimated coefficients\n",
        "print(\"Coefficients determined: \" + str(reg_lassocv.coef_))\n",
        "print(\"Alpha determined: \" + str(reg_lassocv.alpha_))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uvBPoHuklyqe",
        "outputId": "81a666fa-c53e-4e36-f447-fae41af24ead"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "R^2 score for TRAIN: 0.6645896458295648\n",
            "R^2 score for TEST: 0.6738916872104344\n",
            "MSE for TEST: 26.47627318756166\n",
            "Coefficients determined: [ 0.46618103 -0.          0.1815399   0.44525475  0.01399487  0.19196677\n",
            " -0.20870086  0.10736205  0.03286744]\n",
            "Alpha determined: 0.13790237376277725\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Based off the results of the $R^2$ score of the unregularized, Ridge, and Lasso models, we see that there is little difference overall, but the Lasso seems to do marginally worse than OLS.  \n",
        "\n",
        "Notice that Ridge is supposed to add a penalty to features with large weights (or coefficients) in an attempt to reduce overfitting. Lasso is supposed to push features that are insignificant to 0 in an attempt to reduce overfitting.  \n",
        "\n",
        "Lasso is best when there are a few features that contribute a lot to the model. If we continued to push `alpha` bigger, we would see accuracy of Ridge and Lasso decrease as we are essentially reaching an intercept model, which obviously is less accurate than an unregularized linear model. \n",
        "\n",
        "In terms of looking at the bias-variance tradeoff, we also did not note any significant underfitting or overfitting in our analysis, as the training and test scores for the above regression models were very similar. The model also was quite stable when trying different versions (OLS, L1, L2), also showing a proper fit of the model. An MSE of around 26 (RMSE of around 5.14) is also quite solid for our model."
      ],
      "metadata": {
        "id": "F5Tlfq98iOJq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### PCA on Regression Models"
      ],
      "metadata": {
        "id": "Sf0PE9C4eEr3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can try PCA on our linear regression model to see if only picking out prominent features will help our $R^2$ score. However, since PCA is much like Lasso in the sense that some features are \"removed\", we would expect PCA to do worse than the unregularized linear model, because it seems that the features all slightly contribute to the model. However, we can still run PCA on our regression features as a double check!   \n",
        "\n",
        "This is also a way to remove some residual multicollinearity we had in the data"
      ],
      "metadata": {
        "id": "Z-yt8IQPjoRf"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "PCA is not scale-invariant so we need to first standardize the data in order to help us to choose the right number of Principal Components. We also save the explained variance ratios and cumulative explained variance ratios to use for our plot later. "
      ],
      "metadata": {
        "id": "Ny0zqLFvlFnB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Standardizing the features between 0 and 1\n",
        "# create a Scaler object and fit it to your training data\n",
        "scaler = StandardScaler().fit(x_train)\n",
        "x_train_1 = scaler.transform(x_train)\n",
        "\n",
        "#Instantiate and Fit PCA\n",
        "pca = PCA(n_components=9)\n",
        "x_1 = pca.fit_transform(x_train_1)\n",
        "\n",
        "#Save the explained variance ratios\n",
        "explained_variance_ratios = pca.explained_variance_ratio_\n",
        "\n",
        "#Save the CUMULATIVE explained variance ratios\n",
        "cum_evr = np.cumsum(explained_variance_ratios)"
      ],
      "metadata": {
        "id": "xhjOshrRhiff"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now that we've done our preliminary step, let's plot the explained variance ratio to see how many Principal Components we should pick. We're shooting for 85% of our variance to be explained by the number of Principal Components we pick, since this is the value we used in this course."
      ],
      "metadata": {
        "id": "zaeRjTw0lnRl"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot\n",
        "plt.figure(figsize =(16,10))\n",
        "fig, ax = plt.subplots()\n",
        "plot = sns.lineplot(x=range(1,(len(cum_evr)+1)), y=cum_evr, ax = ax)\n",
        "ax.set_xlim(1,10)\n",
        "ax.set_xticks(range(1, 10))\n",
        "plot.axhline(0.85, color = \"green\")\n",
        "\n",
        "# Set label for x-axis\n",
        "plt.xlabel( \"Number of Components\" , size = 10 )\n",
        "  \n",
        "# Set label for y-axis\n",
        "plt.ylabel( \"Cumulative Explained Variance Ratio\" , size = 10 )\n",
        "  \n",
        "# Set title for figure\n",
        "plt.title( \"Cumulative Explained Variance Ratio vs. Number of Components\" , size = 10 )\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 310
        },
        "id": "BLOvj2CmeM_g",
        "outputId": "92d2e92e-abf0-404a-f58e-8629be193324"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1152x720 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "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": [
        "It seems that ~6 is the number of components that explains 85% of the variance. We'll go with 6 in this example. Let's transform our regression features to reflect this."
      ],
      "metadata": {
        "id": "JOeL_L4emXKj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Refit and transform on training with parameter n = 6\n",
        "pca = PCA(n_components=6)\n",
        "x_2 = pca.fit_transform(x_train_1)\n",
        "\n",
        "#Transform on Testing Set\n",
        "x_test_1 = scaler.transform(x_test)\n",
        "x_test_pca = pca.transform(x_test_1)"
      ],
      "metadata": {
        "id": "LGOKhs6Xhzub"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Finding the $R^2$ score of the linear regression after PCA is run. If what we predicted is correct, the $R^2$ score should be lower than the linear regression without PCA."
      ],
      "metadata": {
        "id": "0UhXhoHTmoCq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Initialize model with default parameters and fit it on the training set\n",
        "pca_linear_reg = LinearRegression()\n",
        "pca_linear_reg.fit(x_2, y_train)\n",
        "\n",
        "#predict on test set\n",
        "y_pred = pca_linear_reg.predict(x_test_pca)\n",
        "\n",
        "#finding r-squared\n",
        "pca_score = pca_linear_reg.score(x_2, y_train)\n",
        "print(pca_score)\n",
        "\n",
        "pca_test_score = pca_linear_reg.score(x_test_pca, y_test)\n",
        "print(pca_test_score)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oJhWjetGhog5",
        "outputId": "46beb068-f0c0-4a1c-8304-f5d9a49c4bd6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.6526377142942905\n",
            "0.6652102204578528\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As expected, running PCA on our linear regression model actually seems to yield a similar effect to the Lasso model, and actually slightly lowers the $R^2$ score as well. In the heat map presented earlier in the project, we notice how there are several features that seem to be highly correlated and we would expect PCA to remove those correlated features and only keep features that explain a majority of the variance. "
      ],
      "metadata": {
        "id": "n2woDwbYnKxV"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Neural Network"
      ],
      "metadata": {
        "id": "XflSrvwJEbBn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We also created a simple sklearn Neural Network model to predict the healthiness of food products. The complexity of the model dramatically increased the number of parameters learned, and the introduction of hidden relu activation layers introduced nonlinearity to the model, leading to a higher $R^2$ score. Since there is no inherent reason for nutrition to be linear, this also aims to capture a potential non-linear trend. "
      ],
      "metadata": {
        "id": "o1nD4RHNYKio"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#sklearn implementation\n",
        "from sklearn.neural_network import MLPRegressor\n",
        "num_hidden1 = 8\n",
        "num_hidden2 = 6\n",
        "neural_net = MLPRegressor(hidden_layer_sizes = (num_hidden1, num_hidden2), \n",
        "                          activation = 'relu', solver='adam')\n",
        "neural_net.fit(x_train, y_train)\n",
        "predictions = neural_net.predict(x_test)\n"
      ],
      "metadata": {
        "id": "GwhEfmXPbLEK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#evaluating neural network model\n",
        "nn_train_score = neural_net.score(x_train, y_train)\n",
        "nn_test_score = neural_net.score(x_test, y_test)\n",
        "\n",
        "# Score calculations\n",
        "print(\"R^2 score for TRAIN: \" + str(nn_train_score))\n",
        "print(\"R^2 score for TEST: \" + str(nn_test_score))\n",
        "\n",
        "# calculate the MSE\n",
        "print(\"MSE for TEST: \" + str(mean_squared_error(y_pred, y_test)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5oNrUT_DioH2",
        "outputId": "b708622b-0d28-4e6a-a901-e6593fde2cde"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "R^2 score for TRAIN: 0.9601482783698304\n",
            "R^2 score for TEST: 0.9592074146248808\n",
            "MSE for TEST: 27.181109207974313\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Interestingly enough, we get quite high $R^2$ values, indicating that we can capture a vast majority of the variance in $y$ with our input data. This makes sense given the increased complexity and potential non-linear interactions. However, the MSE is actually slightly higher for the test data! This means that although we capture the trend better, we are likely to be more off (on average!) using the neural network than regression"
      ],
      "metadata": {
        "id": "vJwP7Fg5E-VN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **4.2** Using macronutrients to Predict Healthiness"
      ],
      "metadata": {
        "id": "w9YRB9PlNR_t"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Creating features and test/train split"
      ],
      "metadata": {
        "id": "aSupF7LgLKCq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "For this portion, our only features will be macronutrients, since these are more readily available for unknwon foods and are common ways to measure how \"good\" a food is. "
      ],
      "metadata": {
        "id": "uLhu4aQ6FZz0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Select the necessary features and print the shape\n",
        "X_macro = features[[\"carbohydrates\", \"fat\", \"protein\"]]\n",
        "print(X_macro.shape)\n",
        "print(nutrition_score.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "27wZSgs5HXm9",
        "outputId": "c157d321-ba03-4342-b11e-d08d1784e3e4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(190770, 3)\n",
            "(190770,)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Create a standard test-train split using an 80/20 split"
      ],
      "metadata": {
        "id": "LNj3-pjIFk_D"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Assign appropriate value to seed and conduct 80/20 train-test split with random_state = seed\n",
        "seed = 42\n",
        "y_macro = nutrition_score\n",
        "\n",
        "X_train_macro, X_test_macro, y_train_macro, y_test_macro = train_test_split(X_macro, y_macro, shuffle=True,\n",
        "                                   test_size=0.20, random_state=seed)"
      ],
      "metadata": {
        "id": "xzj8uOBZLJI7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Classification Models: Logistic Regression"
      ],
      "metadata": {
        "id": "_-1OKjG_MC2E"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "As a baseline, let's run a standard logistic regression on the data (without any regularization) to see the accuracies we can get.   \n",
        "\n",
        "Note that there is inherently no class imbalance here because we feature engineered our y-variable to be postive/negative based on comparing it to the median, meaning we have a 50/50 split. Therefore, using accuracies is valid."
      ],
      "metadata": {
        "id": "vBclRj7JF1pw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Initialize model with default parameters and fit it on the training set\n",
        "log_reg = LogisticRegression()\n",
        "log_reg.fit(X_train_macro, y_train_macro)\n",
        "\n",
        "# Use the model to predict on the test set and save these predictions as `y_pred`\n",
        "y_pred = log_reg.predict(X_test_macro)\n",
        "\n",
        "# Find the accuracy and store the value in `log_acc`\n",
        "log_acc = accuracy_score(y_pred, y_test_macro)\n",
        "print(\"Training accuracy: \" + str(log_reg.score(X_train_macro, y_train_macro)))\n",
        "print(\"Testing accuracy: \" + str(log_reg.score(X_test_macro, y_test_macro)))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bAlBbRTeMIUm",
        "outputId": "a133faa5-934f-4ffd-9c72-3e3f6f90203b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training accuracy: 0.8118807988677466\n",
            "Testing accuracy: 0.8124443046600619\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This is pretty good! Also note the similar scores between train/test.     \n",
        "\n",
        "We can also look our confusion matrix and calculate the $F1$ score to ensure there is no weird behavior between classes\n"
      ],
      "metadata": {
        "id": "0GSj5fpYGuqc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.metrics import f1_score\n",
        "\n",
        "# evaluation of Logistic Model\n",
        "print(confusion_matrix(y_test_macro, y_pred))\n",
        "print(f1_score(y_test_macro, y_pred, pos_label = 'healthy'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ueXR3_B30Tqc",
        "outputId": "0395829c-80a0-453d-af88-576a6ab30268"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[15777  3919]\n",
            " [ 3237 15221]]\n",
            "0.8151382071816068\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As expected, both classes behave the same and there are similar amounts of false negatives and false positives. The $F1$ score is satisfactory."
      ],
      "metadata": {
        "id": "F7gXAAvVG4nw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Correlation analysis and PCA if needed"
      ],
      "metadata": {
        "id": "gRESVOa2Nmuy"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "As we move into regularized models and regression, we want to remove correlated features. In EDA and 4.1, we saw correlations between these 3 entities, so we want to remove them via PCA to remove any multicollinearity that will hurt future models."
      ],
      "metadata": {
        "id": "mToEJ5HIG_Lg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_macro.corr() \n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "YYKfg7HFNq_b",
        "outputId": "1bfbe989-2e8a-443d-d696-1b60b56eeede"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               carbohydrates       fat   protein\n",
              "carbohydrates       1.000000 -0.049427 -0.100950\n",
              "fat                -0.049427  1.000000  0.267144\n",
              "protein            -0.100950  0.267144  1.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a3dd63fb-104d-4bc9-b74d-2fcc3d87f490\">\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>carbohydrates</th>\n",
              "      <th>fat</th>\n",
              "      <th>protein</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>carbohydrates</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.049427</td>\n",
              "      <td>-0.100950</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fat</th>\n",
              "      <td>-0.049427</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.267144</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>protein</th>\n",
              "      <td>-0.100950</td>\n",
              "      <td>0.267144</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a3dd63fb-104d-4bc9-b74d-2fcc3d87f490')\"\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-a3dd63fb-104d-4bc9-b74d-2fcc3d87f490 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-a3dd63fb-104d-4bc9-b74d-2fcc3d87f490');\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": 52
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The table has several values with an absolute value > 0.1, so PCA will be good to remove multicolinearity.  \n",
        "\n",
        "Also note that as we move into regression, we need the real-valued scores and not the binarized version."
      ],
      "metadata": {
        "id": "OSbiihmJHNaI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Need real values (nutrition score), not binarized good/bad\n",
        "y_score = df_open_food_ml['nutrition-score-fr_100g']\n",
        "X_train_score, X_test_score, y_train_score, y_test_score = train_test_split(X_macro, y_score, shuffle=True,\n",
        "                                   test_size=0.20, random_state=seed)"
      ],
      "metadata": {
        "id": "TSDaKZdBCWkY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can run PCA in a similar way we have done in the past: standardizing the data, fit_transform to the training data, and doing the same to the test data.   "
      ],
      "metadata": {
        "id": "kHBeuD3SHWzL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#PCA\n",
        "scaler = StandardScaler()\n",
        "\n",
        "scaler.fit(X_train_score)\n",
        "scaled_X_train = scaler.transform(X_train_score)\n",
        "scaled_X_test = scaler.transform(X_test_score)\n",
        "\n",
        "pca = PCA(n_components=scaled_X_train.shape[1])\n",
        "pca_X_train = pca.fit_transform(scaled_X_train)\n",
        "pca_X_test = pca.fit_transform(scaled_X_test)\n",
        "\n",
        "#Variance captured analysis \n",
        "explained_variance_ratios = pca.explained_variance_ratio_\n",
        "cum_evr = np.cumsum(explained_variance_ratios)\n",
        "print(cum_evr)\n",
        "print(pca_X_train.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "07yqv0jeNl7x",
        "outputId": "3bdb7133-6cc8-4406-faf1-4863c06641ab"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.43317333 0.7577854  1.        ]\n",
            "(152616, 3)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Since we only have 3 features and using 2 features does not capture most of the variance (see plot below), we will use all 3 principal components. "
      ],
      "metadata": {
        "id": "2xzsn93EH50o"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Regression models: Elastic Net"
      ],
      "metadata": {
        "id": "FMVBrbYEN_QJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Given the small number of input features, its not worth figuring out feature importance and resultant desired behavior (ex. pushing weights to zero via L1 or shrink via L2), so an elastic net is used for ratios ranging from 0.0 to 1.0.\n",
        "The ElasticNetCV Library allows us to cross-validate in a simple function call by specifying the number of folds to validate with (cv=5)"
      ],
      "metadata": {
        "id": "oQHspdKvfuLd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#PCR to predict the actual score using elastic net (PCA as input)\n",
        "\n",
        "#Use different ratios of L1 (1.0) and L2 (0.0) in the elastic net\n",
        "ratios = [i/100 for i in range(1, 101)]\n",
        "\n",
        "#Use CV\n",
        "elastic = ElasticNetCV(l1_ratio=ratios, cv=5) #Valid range of alphas are automatically tested \n",
        "\n",
        "#Fit and predict\n",
        "model = elastic.fit(pca_X_train, y_train_score)\n",
        "y_pred = model.predict(pca_X_test)"
      ],
      "metadata": {
        "id": "oQbVchcOOAtN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Parameter and MSE analysis for elastic net**"
      ],
      "metadata": {
        "id": "gvHN3gPRC2Iu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "print(\"Alpha chosen: \" + str(model.alpha_))\n",
        "print(\"Coefficients: \" + str(model.coef_))\n",
        "print(\"L1 ratio chosen: \" + str(model.l1_ratio_))\n",
        "print(\"R^2 Score for TRAIN: \" + str(model.score(pca_X_train, y_train_score)))\n",
        "print(\"R^2 Score for TEST: \" + str(model.score(pca_X_test, y_test_score)))\n",
        "print(\"Test MSE: \" + str(mean_squared_error(y_pred, y_test_score)))\n",
        "print(\"Train MSE: \" + str(mean_squared_error(model.predict(pca_X_train), y_train_score)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JfKcR3_CC0mv",
        "outputId": "31ee5889-9ed4-42ed-9b34-e93f57486212"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Alpha chosen: 0.0040486647687908925\n",
            "Coefficients: [ 2.35998276  4.18407394 -3.34148818]\n",
            "L1 ratio chosen: 1.0\n",
            "R^2 Score for TRAIN: 0.39718016935361555\n",
            "R^2 Score for TEST: 0.3994992712505223\n",
            "Test MSE: 48.753805776059764\n",
            "Train MSE: 49.174478352128375\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Overall, we see the parameters chosen are a very low alpha and a complete focus on L1 regularization. The coefficients are all significantly non-zero, but note they are for the principal components and not for the macronutrient themselves, so there is not much interpretability here. \n",
        "\n",
        "The $R^2$ values are quite low here, only about 0.4 (similar again for test and train!). The MSE is not bad at all given that we only have 3 features, but is still quite a bit higher than the ~25 we were getting in 4.1 (but again, similar between test and train indicating lack of overfit!). Let's try something more complex."
      ],
      "metadata": {
        "id": "EXsoIUoxJNta"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Regression Models: Random Forests"
      ],
      "metadata": {
        "id": "yraRTGX_JG-Y"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "While logistic regression does a decent job at deciding a valid cut-off, our performance increase from regression to neural network in 4.1 shows inherent nonlinearities within the data.   \n",
        "\n",
        "Here, we use RandomForests to regress our data to add more complexity."
      ],
      "metadata": {
        "id": "k5z_7NhjH7Ps"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can use cross-validation to pick the ideal n_estimators and max_depth for random forests. **This takes ~15 minutes to run because of the loops needed for cross-validation that produces 9 iterations!**\n",
        "\n",
        "Ideally, more combinations n_estimators would be tested, but this causes too much time/computing power to run effectively."
      ],
      "metadata": {
        "id": "QJv0nKlZWRYW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "random_seed = 42\n",
        "depths = [1, 2, 3] #Only 3 features \n",
        "n_estimators = [200*i for i in range(1, 4)]\n",
        "mse = []\n",
        "import sklearn\n",
        "print(sklearn.metrics.SCORERS.keys())\n",
        "for depth in depths: \n",
        "  for n_est in n_estimators: \n",
        "    print(\"next iter\") #Keep track when the model is finished\n",
        "    rf = RandomForestRegressor(random_state=random_seed, max_depth=depth, n_estimators=n_est)\n",
        "    cv = KFold(n_splits=5).get_n_splits(pca_X_train, y_train_score)\n",
        "    scores = cross_val_score(rf, pca_X_train, y_train_score, cv=cv, scoring='neg_mean_squared_error')\n",
        "    mse.append(abs(np.mean(scores)))\n",
        "print(mse)\n",
        "print(np.argmin(mse))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PK8pQZ06IHFR",
        "outputId": "23696f57-25e1-4d52-bab0-067d2d912383"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "dict_keys(['explained_variance', 'r2', 'max_error', 'neg_median_absolute_error', 'neg_mean_absolute_error', 'neg_mean_absolute_percentage_error', 'neg_mean_squared_error', 'neg_mean_squared_log_error', 'neg_root_mean_squared_error', 'neg_mean_poisson_deviance', 'neg_mean_gamma_deviance', 'accuracy', 'top_k_accuracy', 'roc_auc', 'roc_auc_ovr', 'roc_auc_ovo', 'roc_auc_ovr_weighted', 'roc_auc_ovo_weighted', 'balanced_accuracy', 'average_precision', 'neg_log_loss', 'neg_brier_score', 'adjusted_rand_score', 'rand_score', 'homogeneity_score', 'completeness_score', 'v_measure_score', 'mutual_info_score', 'adjusted_mutual_info_score', 'normalized_mutual_info_score', 'fowlkes_mallows_score', 'precision', 'precision_macro', 'precision_micro', 'precision_samples', 'precision_weighted', 'recall', 'recall_macro', 'recall_micro', 'recall_samples', 'recall_weighted', 'f1', 'f1_macro', 'f1_micro', 'f1_samples', 'f1_weighted', 'jaccard', 'jaccard_macro', 'jaccard_micro', 'jaccard_samples', 'jaccard_weighted'])\n",
            "next iter\n",
            "next iter\n",
            "next iter\n",
            "next iter\n",
            "next iter\n",
            "next iter\n",
            "next iter\n",
            "next iter\n",
            "next iter\n",
            "[60.98976875004363, 60.98888470577565, 60.98653727747588, 50.45853154033226, 50.45011561366463, 50.45011118065855, 42.55499229281547, 42.542526911786766, 42.541854363826864]\n",
            "8\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The results from running cell (if you don't have 15 minutes) are: \n",
        "\n",
        "\n",
        "```\n",
        "MSE: [59.09639375429904, 59.102198417439126, 59.10936757856681, 49.30632169462469, 49.36827237522546, 49.40105322143889, 40.79416782775635, 40.84685514089898, 40.87357334867751]\n",
        "Argmax index: 2\n",
        "Argmin index: 6\n",
        "```\n",
        "\n",
        "\n",
        "From cross-validation, we get that the idea hyperparameter set is n_est = 200, max_depth = 3. We can train this model on the whole training set and test it on the test set to get the corresponding score and mean squared error."
      ],
      "metadata": {
        "id": "QbUdzx-xWcw4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#[59.09639375429904, 59.102198417439126, 59.10936757856681, 49.30632169462469, 49.36827237522546, 49.40105322143889, 40.79416782775635, 40.84685514089898, 40.87357334867751]\n",
        "#argmin index -> 6\n",
        "\n",
        "depth=3\n",
        "n_est=200\n",
        "random_seed = 42\n",
        "rf = RandomForestRegressor(random_state=random_seed, max_depth=depth, n_estimators=n_est)\n",
        "rf.fit(pca_X_train, y_train_score)\n",
        "y_pred = rf.predict(pca_X_test)\n"
      ],
      "metadata": {
        "id": "aI8tQUrIWqQH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Print out metrics (without having to re-run!)"
      ],
      "metadata": {
        "id": "uD8jr9wNLogo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"R^2 Score for TRAIN: \" + str(rf.score(pca_X_train, y_train_score)))\n",
        "print(\"R^2 Score for TEST: \" + str(rf.score(pca_X_test, y_test_score)))\n",
        "print(\"Test MSE: \" + str(mean_squared_error(y_pred, y_test_score)))\n",
        "print(\"Train MSE: \" + str(mean_squared_error(rf.predict(pca_X_train), y_train_score)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fGtZN0rpLn8a",
        "outputId": "6a2e27d7-0276-47f8-b8d3-319191719e1b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "R^2 Score for TRAIN: 0.4788927342010001\n",
            "R^2 Score for TEST: 0.4767730600160959\n",
            "Test MSE: 42.48005603906526\n",
            "Train MSE: 42.50885033707778\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "These results are far better than our elastic net metrics. We obtain an $R^2$ value of nearly 0.5 (as opposed to ~0.39) and our mean-squared-error has decreased to around 42 (as opposed to ~50). \n",
        "\n",
        "Again, note similar values between train and test indicating lack of overfitting.   \n",
        "\n",
        "Overall: We are off by about 6-7 (on average) for each prediction, which is worse than our 4.1 model, but with a LOT less information. Only given macronutrient information (carbs, fat, protein), we built a pretty accurate classifier and regressor to calculate its nutrition score. "
      ],
      "metadata": {
        "id": "b2wukP7pioGl"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Side note: why no NNs:(?**   \n",
        "A: Only 3 input features makes this a weak candidate for complex models like neural networks and will likely cause overfitting."
      ],
      "metadata": {
        "id": "WdrpL81NOJjt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **5** Conclusion: Summary of al analyses and findings"
      ],
      "metadata": {
        "id": "i97YPF3svAMt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Through the Open Food Facts Kaggle dataset, we are able to clean, explore, analyze, and model food facts data through various lenses. After mounting the Kaggle dataset and some data cleaning, we were able to do some exploratory data analysis on the dataset. As a first step, we wanted to get an idea of which countries the majority of the food products were coming from, as this might have an influence on nutrition and which cuisines were represented -- US and France had the highest product counts. We also found the countries with healthiest foods in the dataset, based on nutrition score, with Pakistan at the top of the list. Next, we looked at the ingredients list column and broke it down. The motivation for this was to see the most common ingredients across various products in the dataset. We then ordered and found the top ingredients, some of which include salt and oil. To visualize this, we put it in a wordcloud through nltk. We then looked at the nutrients section of the data and plotted boxplots of each nutrient present (with outliers removed as it hurt visualization), and found that most foods tend to have low-to-none values for most nutrients but the macronutrients have high variability. Creating another boxplot set for the macronutrients showed higher variability in carbohydrates than protein and fat.\n",
        "\n",
        "In terms of modeling, we mainly wanted to regress on a product’s various features to predict its nutrition score. Firstly, we noticed that there is multicolinearity between some features, such as salt and sodium, so we removed features that showed high colinearity. To achieve regression on nutrition score, we ran an unregularized linear regression, a PCA linear regression, a LASSO linear regression (with cross-val), and a Ridge linear regression (with cross-val). We noticed that the linear and ridge regressions (roughly 0.67 R-squared score) had similar R-squared scores while LASSO and PCA linear regression (roughly 0.62 R-squared score) performed not as well as the previous models. We suspect that this is because the features all contribute to the model to some extent where removing features or forcing features to zero would actually lower accuracy. We also ran a simple neural network model to predict nutrition score with the same features using the sklearn library, with two hidden layers. The model yielded a surprisingly high accuracy score (approx 0.96 R-squared score) on both the training and test sets; this may be due to the nonlinearity the relu layers of the model introduced, in contrast to the previous few models. \n",
        "Our second goal was to be able to classify products exclusively on macronutrient features to predict whether a product is “healthy” or not based on a binary version of nutrition score. Using simple logistic regression as a baseline, we were able to get an accuracy of above 0.81 using only fat, protein, and carbs. We then analyzed the data and saw multicollinerity, so PCA was run with all principal components kept (3). Given the small number of features, elastic net was used since it was hard to determine whether pushing weights to zero (L1) or holistically reducing them (L2) is ideal. The elastic net (with cross-val) got an R^2 score of about 0.39 with an MSE of around 50 - worse than our previous regressors, but using a lot less data. To increase model complexity, we used a RandomForestRegressor with cross-validation and got an R^2 score of ~0.48 and an MSE of around 42.5. Using only 3 features, these values are incredibly impressive at calculating various foods' nutrition scores.\n",
        "\n",
        "Overall, we were able to conduct some exploratory data analysis and modeling on the Open Food Facts dataset. Moving forward, we can conduct more exploratory data analysis on other variables within the dataset, such as exploring the URL and conducting web scraping to gain more information about each product. It would also be helpful to look more at the effectiveness of various other models, such as boosting and even deep learning models. \n",
        "\n",
        "\n",
        "We learned a lot about visualization, correlations, statistics, machine learning classification, and machine learning regression during this project. We hope to apply these models into our own lives and predict the healthiness of the foods we snack on!"
      ],
      "metadata": {
        "id": "GzlOdZYuvGJH"
      }
    }
  ]
}