anime_recommendation_system.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "anime-recommendation-system.ipynb",
      "provenance": [],
      "machine_shape": "hm",
      "authorship_tag": "ABX9TyNQn11marK+JYjoiF/Xkxu2",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/earthlovebpt/myprojects/blob/main/anime_recommendation_system.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 0. Setup"
      ],
      "metadata": {
        "id": "kWn87K_bTaYC"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J3N87khIJaZH",
        "outputId": "c9b2e1b6-31e0-489a-94fa-08e2b57ffc9a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "! cp /content/drive/MyDrive/projects/anime-recommedation-system/data.zip data.zip\n",
        "! unzip -qq /content/data.zip"
      ],
      "metadata": {
        "id": "9nBlDMMMJrhk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import random\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "plt.style.use('ggplot')\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from tqdm.notebook import tqdm\n",
        "\n",
        "\n",
        "from tensorflow.keras import layers, Model\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau, CSVLogger\n",
        "from tensorflow.keras import optimizers\n",
        "from tensorflow.keras.models import load_model"
      ],
      "metadata": {
        "id": "dp2XV7-ILH9d"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 1. EDA"
      ],
      "metadata": {
        "id": "ZIEDOqXETd_d"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Read the csv file.\n",
        "df = pd.read_csv('/content/animelist.csv', nrows=5e7)\n",
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "h5S20rRcLIOO",
        "outputId": "068a05cd-f52f-436e-bca3-e24a2cc9d8ee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          user_id  anime_id  rating  watching_status  watched_episodes\n",
              "0               0        67       9                1                 1\n",
              "1               0      6702       7                1                 4\n",
              "2               0       242      10                1                 4\n",
              "3               0      4898       0                1                 1\n",
              "4               0        21      10                1                 0\n",
              "...           ...       ...     ...              ...               ...\n",
              "49999995   162169      5178       0                6                 0\n",
              "49999996   162169     25183       0                6                 0\n",
              "49999997   162169      1571       0                6                 0\n",
              "49999998   162169     21431       0                6                 0\n",
              "49999999   162169     17895       0                6                 0\n",
              "\n",
              "[50000000 rows x 5 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f3417d3e-65ae-4a44-bd48-6c0beffab8ea\">\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>user_id</th>\n",
              "      <th>anime_id</th>\n",
              "      <th>rating</th>\n",
              "      <th>watching_status</th>\n",
              "      <th>watched_episodes</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>67</td>\n",
              "      <td>9</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>6702</td>\n",
              "      <td>7</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>242</td>\n",
              "      <td>10</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>4898</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>21</td>\n",
              "      <td>10</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999995</th>\n",
              "      <td>162169</td>\n",
              "      <td>5178</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999996</th>\n",
              "      <td>162169</td>\n",
              "      <td>25183</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999997</th>\n",
              "      <td>162169</td>\n",
              "      <td>1571</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999998</th>\n",
              "      <td>162169</td>\n",
              "      <td>21431</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999999</th>\n",
              "      <td>162169</td>\n",
              "      <td>17895</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>50000000 rows × 5 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f3417d3e-65ae-4a44-bd48-6c0beffab8ea')\"\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-f3417d3e-65ae-4a44-bd48-6c0beffab8ea 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-f3417d3e-65ae-4a44-bd48-6c0beffab8ea');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.read_csv('/content/watching_status.csv')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "ULZzkj0cyz8P",
        "outputId": "20c1488b-fda1-406d-bbb6-4d167421e057"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   status         description\n",
              "0       1  Currently Watching\n",
              "1       2           Completed\n",
              "2       3             On Hold\n",
              "3       4             Dropped\n",
              "4       6       Plan to Watch"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e474eb53-6b34-4f02-b041-a181f25606eb\">\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>status</th>\n",
              "      <th>description</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Currently Watching</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>Completed</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>On Hold</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>Dropped</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6</td>\n",
              "      <td>Plan to Watch</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e474eb53-6b34-4f02-b041-a181f25606eb')\"\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-e474eb53-6b34-4f02-b041-a181f25606eb 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-e474eb53-6b34-4f02-b041-a181f25606eb');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = df[df['watching_status'].isin([0,1,2,3,4,5,6])]"
      ],
      "metadata": {
        "id": "4-s9ABsdVnUH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot the histogram.\n",
        "plt.figure(figsize=(15,8))\n",
        "sns.countplot(data=df, x='rating', hue='watching_status')\n",
        "plt.title('Histogram of anime rating')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        },
        "id": "f3YDlx8S3296",
        "outputId": "8b3e9d8a-d0bd-408c-ed5a-fb7e3cdc78c4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "From the histogram, rating on animes cannot be used directly since the animes that are planned to watch (status=6) usually get rating 0. Consequently, we exlude the rating 0."
      ],
      "metadata": {
        "id": "v6RDz4x55E9D"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Exclude the plan-to-watch anime record.\n",
        "df = df[df['rating']!=0]\n",
        "print(f'The number of left data: {df.shape[0]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w54bd_Fy6MWn",
        "outputId": "26d2a2b2-3e19-4829-fdc7-1673337ec0d6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The number of left data: 28535793\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Scale rating into range from 0 to 1.\n",
        "df['scaled_rating'] = df['rating']/10\n",
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "QGgmFjnHBFH5",
        "outputId": "f45cab5c-db92-408d-fd32-7fea06188b29"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          user_id  anime_id  rating  watching_status  watched_episodes  \\\n",
              "0               0        67       9                1                 1   \n",
              "1               0      6702       7                1                 4   \n",
              "2               0       242      10                1                 4   \n",
              "4               0        21      10                1                 0   \n",
              "5               0        24       9                1                 5   \n",
              "...           ...       ...     ...              ...               ...   \n",
              "49999955   162169      3712       6                2                12   \n",
              "49999956   162169     11837       7                2                13   \n",
              "49999957   162169     14075       6                2                24   \n",
              "49999959   162169      6702       6                3               102   \n",
              "49999969   162169     34933       6                4                 4   \n",
              "\n",
              "          scaled_rating  \n",
              "0                   0.9  \n",
              "1                   0.7  \n",
              "2                   1.0  \n",
              "4                   1.0  \n",
              "5                   0.9  \n",
              "...                 ...  \n",
              "49999955            0.6  \n",
              "49999956            0.7  \n",
              "49999957            0.6  \n",
              "49999959            0.6  \n",
              "49999969            0.6  \n",
              "\n",
              "[28535793 rows x 6 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-725ddc17-3310-4c29-ad05-1aa2c334081a\">\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>user_id</th>\n",
              "      <th>anime_id</th>\n",
              "      <th>rating</th>\n",
              "      <th>watching_status</th>\n",
              "      <th>watched_episodes</th>\n",
              "      <th>scaled_rating</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>67</td>\n",
              "      <td>9</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0.9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>6702</td>\n",
              "      <td>7</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>0.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>242</td>\n",
              "      <td>10</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>21</td>\n",
              "      <td>10</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0</td>\n",
              "      <td>24</td>\n",
              "      <td>9</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>0.9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999955</th>\n",
              "      <td>162169</td>\n",
              "      <td>3712</td>\n",
              "      <td>6</td>\n",
              "      <td>2</td>\n",
              "      <td>12</td>\n",
              "      <td>0.6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999956</th>\n",
              "      <td>162169</td>\n",
              "      <td>11837</td>\n",
              "      <td>7</td>\n",
              "      <td>2</td>\n",
              "      <td>13</td>\n",
              "      <td>0.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999957</th>\n",
              "      <td>162169</td>\n",
              "      <td>14075</td>\n",
              "      <td>6</td>\n",
              "      <td>2</td>\n",
              "      <td>24</td>\n",
              "      <td>0.6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999959</th>\n",
              "      <td>162169</td>\n",
              "      <td>6702</td>\n",
              "      <td>6</td>\n",
              "      <td>3</td>\n",
              "      <td>102</td>\n",
              "      <td>0.6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49999969</th>\n",
              "      <td>162169</td>\n",
              "      <td>34933</td>\n",
              "      <td>6</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>0.6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>28535793 rows × 6 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-725ddc17-3310-4c29-ad05-1aa2c334081a')\"\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-725ddc17-3310-4c29-ad05-1aa2c334081a 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-725ddc17-3310-4c29-ad05-1aa2c334081a');\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": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot the histogram.\n",
        "plt.figure(figsize=(15,8))\n",
        "sns.countplot(data=df, x='rating')\n",
        "plt.title('Histogram of anime rating without rating 0')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        },
        "id": "r_RyOC50JK1m",
        "outputId": "504b3e0f-cff0-49c3-ec92-f0531ebd2ee4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 2. Prepare the data"
      ],
      "metadata": {
        "id": "G2qH3uwLoYEi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Encoding the user_id.\n",
        "user_list = df[\"user_id\"].unique().tolist()\n",
        "user_encoded = {user:idx for idx,user in enumerate(user_list)}\n",
        "user_decoded = {idx:user for idx,user in enumerate(user_list)}\n",
        "df[\"user\"] = df[\"user_id\"].map(user_encoded)\n",
        "n_users = len(user_encoded)\n",
        "\n",
        "# Encoding the anime_id.\n",
        "anime_list = df[\"anime_id\"].unique().tolist()\n",
        "anime_encoded = {anime:idx for idx,anime in enumerate(anime_list)}\n",
        "anime_decoded = {idx:anime for idx,anime in enumerate(anime_list)}\n",
        "df[\"anime\"] = df[\"anime_id\"].map(anime_encoded)\n",
        "n_animes = len(anime_encoded)"
      ],
      "metadata": {
        "id": "0fhyN2DGkmh0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Split data into training data(60%), evaluating data(20%), and testing data(20%).\n",
        "train_df, new_df = train_test_split(df, test_size=0.4, random_state=0)\n",
        "test_df, eval_df = train_test_split(new_df, test_size=0.5, random_state=0)"
      ],
      "metadata": {
        "id": "5QVLQ4KcmE0I"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print('The number of training data:   ',train_df.shape[0])\n",
        "print('The number of evaluating data: ',eval_df.shape[0])\n",
        "print('The number of testing data:    ',test_df.shape[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7wQfIQPUmrI7",
        "outputId": "dcee19ae-9409-4cbb-c14a-77498d876534"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The number of training data:    17121475\n",
            "The number of evaluating data:  5707159\n",
            "The number of testing data:     5707159\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_train = [train_df['user'].values, train_df['anime'].values]\n",
        "y_train = train_df['scaled_rating'].values.reshape(-1,1)\n",
        "\n",
        "\n",
        "X_eval = [eval_df['user'].values, eval_df['anime'].values]\n",
        "y_eval = eval_df['scaled_rating'].values.reshape(-1,1)\n",
        "\n",
        "\n",
        "X_test = [test_df['user'].values, test_df['anime'].values]\n",
        "y_test = test_df['scaled_rating'].values.reshape(-1,1)"
      ],
      "metadata": {
        "id": "5Z8Kcxlcl5Em"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 3. Create Model"
      ],
      "metadata": {
        "id": "xk35XfQoswaz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def create_model(embedding_size=128):\n",
        "  user = layers.Input(name='user', shape = (1,))\n",
        "  user_embedding = layers.Embedding(name='user_embedding', input_dim=n_users, output_dim=embedding_size)(user)\n",
        "\n",
        "  anime = layers.Input(name='anime', shape = (1,))\n",
        "  anime_embedding = layers.Embedding(name='anime_embedding', input_dim=n_animes, output_dim=embedding_size)(anime)\n",
        "\n",
        "\n",
        "  x = layers.Dot(name = 'dot_product', normalize = True, axes = 2)([user_embedding, anime_embedding])\n",
        "  x = layers.Flatten()(x)\n",
        "  output = layers.Dense(1, activation='sigmoid')(x)\n",
        "\n",
        "  model = Model(inputs=[user, anime], outputs=output)\n",
        "  model.compile(loss='mean_absolute_error', metrics=['mae','mse'], optimizer=optimizers.Adam(lr=0.0002))\n",
        "  \n",
        "  return model"
      ],
      "metadata": {
        "id": "FH7A53uQqzQD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model = create_model()\n",
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bPtNLmHOss_S",
        "outputId": "65b94321-6ba8-4291-dae8-4bc5ca2558af"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_2\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " user (InputLayer)              [(None, 1)]          0           []                               \n",
            "                                                                                                  \n",
            " anime (InputLayer)             [(None, 1)]          0           []                               \n",
            "                                                                                                  \n",
            " user_embedding (Embedding)     (None, 1, 128)       18417920    ['user[0][0]']                   \n",
            "                                                                                                  \n",
            " anime_embedding (Embedding)    (None, 1, 128)       2169984     ['anime[0][0]']                  \n",
            "                                                                                                  \n",
            " dot_product (Dot)              (None, 1, 1)         0           ['user_embedding[0][0]',         \n",
            "                                                                  'anime_embedding[0][0]']        \n",
            "                                                                                                  \n",
            " flatten_2 (Flatten)            (None, 1)            0           ['dot_product[0][0]']            \n",
            "                                                                                                  \n",
            " dense_2 (Dense)                (None, 1)            2           ['flatten_2[0][0]']              \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 20,587,906\n",
            "Trainable params: 20,587,906\n",
            "Non-trainable params: 0\n",
            "__________________________________________________________________________________________________\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Save the best model.\n",
        "checkpoint = ModelCheckpoint(filepath='best_model.h5',\n",
        "                             save_best_only=True,\n",
        "                             verbose=1,\n",
        "                             mode='min',\n",
        "                             moniter='val_loss'\n",
        "                            )\n",
        "\n",
        "# Reduce lr when val_loss does not decrease.\n",
        "reduce_lr = ReduceLROnPlateau(monitor='val_loss', \n",
        "                              factor=0.5, \n",
        "                              patience=3, \n",
        "                              verbose=1, \n",
        "                              min_delta=0.001\n",
        "                             )\n",
        "\n",
        "# Save log file.\n",
        "csv_logger = CSVLogger('training.log')\n",
        "\n",
        "callbacks = [checkpoint, reduce_lr, csv_logger]"
      ],
      "metadata": {
        "id": "TKHrIpWDrtcI"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Model training\n",
        "hist = model.fit(\n",
        "    x=X_train,\n",
        "    y=y_train,\n",
        "    batch_size=10000,\n",
        "    epochs=20,\n",
        "    verbose=1,\n",
        "    validation_data=(X_eval, y_eval),\n",
        "    validation_batch_size=5000,\n",
        "    callbacks=callbacks\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zmvRVyUeqGCc",
        "outputId": "db6194d4-7c61-4cde-cecb-7be68a1f2806"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.2097 - mae: 0.2097 - mse: 0.0615\n",
            "Epoch 1: val_loss improved from inf to 0.13358, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 19s 10ms/step - loss: 0.2096 - mae: 0.2096 - mse: 0.0614 - val_loss: 0.1336 - val_mae: 0.1336 - val_mse: 0.0285 - lr: 2.0000e-04\n",
            "Epoch 2/20\n",
            "1711/1713 [============================>.] - ETA: 0s - loss: 0.1170 - mae: 0.1170 - mse: 0.0229\n",
            "Epoch 2: val_loss improved from 0.13358 to 0.10655, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 10ms/step - loss: 0.1170 - mae: 0.1170 - mse: 0.0229 - val_loss: 0.1066 - val_mae: 0.1066 - val_mse: 0.0197 - lr: 2.0000e-04\n",
            "Epoch 3/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.1017 - mae: 0.1017 - mse: 0.0181\n",
            "Epoch 3: val_loss improved from 0.10655 to 0.10016, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 10ms/step - loss: 0.1017 - mae: 0.1017 - mse: 0.0181 - val_loss: 0.1002 - val_mae: 0.1002 - val_mse: 0.0175 - lr: 2.0000e-04\n",
            "Epoch 4/20\n",
            "1712/1713 [============================>.] - ETA: 0s - loss: 0.0957 - mae: 0.0957 - mse: 0.0162\n",
            "Epoch 4: val_loss improved from 0.10016 to 0.09593, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 11ms/step - loss: 0.0957 - mae: 0.0957 - mse: 0.0162 - val_loss: 0.0959 - val_mae: 0.0959 - val_mse: 0.0163 - lr: 2.0000e-04\n",
            "Epoch 5/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.0904 - mae: 0.0904 - mse: 0.0150\n",
            "Epoch 5: val_loss improved from 0.09593 to 0.09304, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 10ms/step - loss: 0.0904 - mae: 0.0904 - mse: 0.0150 - val_loss: 0.0930 - val_mae: 0.0930 - val_mse: 0.0156 - lr: 2.0000e-04\n",
            "Epoch 6/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.0861 - mae: 0.0861 - mse: 0.0140\n",
            "Epoch 6: val_loss improved from 0.09304 to 0.09146, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 10ms/step - loss: 0.0861 - mae: 0.0861 - mse: 0.0140 - val_loss: 0.0915 - val_mae: 0.0915 - val_mse: 0.0152 - lr: 2.0000e-04\n",
            "Epoch 7/20\n",
            "1708/1713 [============================>.] - ETA: 0s - loss: 0.0822 - mae: 0.0822 - mse: 0.0131\n",
            "Epoch 7: val_loss improved from 0.09146 to 0.09049, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 10ms/step - loss: 0.0822 - mae: 0.0822 - mse: 0.0131 - val_loss: 0.0905 - val_mae: 0.0905 - val_mse: 0.0149 - lr: 2.0000e-04\n",
            "Epoch 8/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.0784 - mae: 0.0784 - mse: 0.0122\n",
            "Epoch 8: val_loss improved from 0.09049 to 0.08996, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 10ms/step - loss: 0.0784 - mae: 0.0784 - mse: 0.0122 - val_loss: 0.0900 - val_mae: 0.0900 - val_mse: 0.0148 - lr: 2.0000e-04\n",
            "Epoch 9/20\n",
            "1709/1713 [============================>.] - ETA: 0s - loss: 0.0749 - mae: 0.0749 - mse: 0.0115\n",
            "Epoch 9: val_loss improved from 0.08996 to 0.08974, saving model to best_model.h5\n",
            "1713/1713 [==============================] - 18s 10ms/step - loss: 0.0749 - mae: 0.0749 - mse: 0.0115 - val_loss: 0.0897 - val_mae: 0.0897 - val_mse: 0.0147 - lr: 2.0000e-04\n",
            "Epoch 10/20\n",
            "1709/1713 [============================>.] - ETA: 0s - loss: 0.0716 - mae: 0.0716 - mse: 0.0108\n",
            "Epoch 10: val_loss did not improve from 0.08974\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0716 - mae: 0.0716 - mse: 0.0108 - val_loss: 0.0898 - val_mae: 0.0898 - val_mse: 0.0147 - lr: 2.0000e-04\n",
            "Epoch 11/20\n",
            "1713/1713 [==============================] - ETA: 0s - loss: 0.0687 - mae: 0.0687 - mse: 0.0101\n",
            "Epoch 11: val_loss did not improve from 0.08974\n",
            "\n",
            "Epoch 11: ReduceLROnPlateau reducing learning rate to 9.999999747378752e-05.\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0687 - mae: 0.0687 - mse: 0.0101 - val_loss: 0.0900 - val_mae: 0.0900 - val_mse: 0.0148 - lr: 2.0000e-04\n",
            "Epoch 12/20\n",
            "1709/1713 [============================>.] - ETA: 0s - loss: 0.0648 - mae: 0.0648 - mse: 0.0094\n",
            "Epoch 12: val_loss did not improve from 0.08974\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0648 - mae: 0.0648 - mse: 0.0094 - val_loss: 0.0900 - val_mae: 0.0900 - val_mse: 0.0148 - lr: 1.0000e-04\n",
            "Epoch 13/20\n",
            "1709/1713 [============================>.] - ETA: 0s - loss: 0.0628 - mae: 0.0628 - mse: 0.0091\n",
            "Epoch 13: val_loss did not improve from 0.08974\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0628 - mae: 0.0628 - mse: 0.0091 - val_loss: 0.0903 - val_mae: 0.0903 - val_mse: 0.0148 - lr: 1.0000e-04\n",
            "Epoch 14/20\n",
            "1709/1713 [============================>.] - ETA: 0s - loss: 0.0613 - mae: 0.0613 - mse: 0.0088\n",
            "Epoch 14: val_loss did not improve from 0.08974\n",
            "\n",
            "Epoch 14: ReduceLROnPlateau reducing learning rate to 4.999999873689376e-05.\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0613 - mae: 0.0613 - mse: 0.0088 - val_loss: 0.0906 - val_mae: 0.0906 - val_mse: 0.0149 - lr: 1.0000e-04\n",
            "Epoch 15/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.0592 - mae: 0.0592 - mse: 0.0085\n",
            "Epoch 15: val_loss did not improve from 0.08974\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0592 - mae: 0.0592 - mse: 0.0085 - val_loss: 0.0907 - val_mae: 0.0907 - val_mse: 0.0149 - lr: 5.0000e-05\n",
            "Epoch 16/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.0583 - mae: 0.0583 - mse: 0.0083\n",
            "Epoch 16: val_loss did not improve from 0.08974\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0583 - mae: 0.0583 - mse: 0.0083 - val_loss: 0.0909 - val_mae: 0.0909 - val_mse: 0.0150 - lr: 5.0000e-05\n",
            "Epoch 17/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.0575 - mae: 0.0575 - mse: 0.0082\n",
            "Epoch 17: val_loss did not improve from 0.08974\n",
            "\n",
            "Epoch 17: ReduceLROnPlateau reducing learning rate to 2.499999936844688e-05.\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0575 - mae: 0.0575 - mse: 0.0082 - val_loss: 0.0911 - val_mae: 0.0911 - val_mse: 0.0150 - lr: 5.0000e-05\n",
            "Epoch 18/20\n",
            "1709/1713 [============================>.] - ETA: 0s - loss: 0.0564 - mae: 0.0564 - mse: 0.0081\n",
            "Epoch 18: val_loss did not improve from 0.08974\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0564 - mae: 0.0564 - mse: 0.0081 - val_loss: 0.0912 - val_mae: 0.0912 - val_mse: 0.0151 - lr: 2.5000e-05\n",
            "Epoch 19/20\n",
            "1709/1713 [============================>.] - ETA: 0s - loss: 0.0559 - mae: 0.0559 - mse: 0.0080\n",
            "Epoch 19: val_loss did not improve from 0.08974\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0559 - mae: 0.0559 - mse: 0.0080 - val_loss: 0.0913 - val_mae: 0.0913 - val_mse: 0.0151 - lr: 2.5000e-05\n",
            "Epoch 20/20\n",
            "1710/1713 [============================>.] - ETA: 0s - loss: 0.0555 - mae: 0.0555 - mse: 0.0080\n",
            "Epoch 20: val_loss did not improve from 0.08974\n",
            "\n",
            "Epoch 20: ReduceLROnPlateau reducing learning rate to 1.249999968422344e-05.\n",
            "1713/1713 [==============================] - 17s 10ms/step - loss: 0.0555 - mae: 0.0555 - mse: 0.0080 - val_loss: 0.0914 - val_mae: 0.0914 - val_mse: 0.0151 - lr: 2.5000e-05\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_history():\n",
        "  plt.figure(figsize=(10,5))\n",
        "  plt.plot(hist.history['loss'])\n",
        "  plt.plot(hist.history['val_loss'])\n",
        "  plt.title('Model Loss')\n",
        "  plt.xlabel('Epochs')\n",
        "  plt.ylabel('Loss')\n",
        "  plt.legend(['train', 'val'])\n",
        "  plt.show()\n",
        "\n",
        "plot_history()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 354
        },
        "id": "vU092oouslrt",
        "outputId": "b2277f48-acff-467d-c9c3-6a1366caba70"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the model on the test set.\n",
        "best_model = load_model('/content/best_model.h5')\n",
        "best_model.evaluate(X_test, y_test, batch_size=2000)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TLdV5Qcs_luI",
        "outputId": "dd8c4b0d-63c5-4e95-8d37-a30fe59b970f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2854/2854 [==============================] - 6s 2ms/step - loss: 0.0898 - mae: 0.0898 - mse: 0.0147\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.08976627141237259, 0.08976627141237259, 0.014703143388032913]"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 4. Recommendation system"
      ],
      "metadata": {
        "id": "rQ6AZUu-tG5L"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 4.1 Recommend similar animes"
      ],
      "metadata": {
        "id": "rLlR8Dh509a9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Get weights from the embedding layer.\n",
        "def extract_weights(name, model):\n",
        "    weight_layer = model.get_layer(name)\n",
        "    weights = weight_layer.get_weights()[0]\n",
        "    weights = weights / np.linalg.norm(weights, axis = 1).reshape((-1, 1))  # Normalize the weights\n",
        "    return weights\n",
        "\n",
        "anime_weights = extract_weights('anime_embedding', best_model)\n",
        "user_weights = extract_weights('user_embedding', best_model)"
      ],
      "metadata": {
        "id": "iw5nNh1GeTkH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "synopsis = pd.read_csv('/content/anime.csv')\n",
        "name_list = []\n",
        "\n",
        "# Use English name to be shown in recommendation.\n",
        "for row in synopsis[['Name', 'English name']].iterrows():\n",
        "  if row[1][1] != 'Unknown':\n",
        "    name_list.append(row[1][1])\n",
        "  else:\n",
        "    name_list.append(row[1][0])\n",
        "synopsis['show_name'] = name_list "
      ],
      "metadata": {
        "id": "r7YjWuk3gpuh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "synopsis = synopsis[['MAL_ID', 'show_name','Genres']]\n",
        "synopsis"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "jShHHGWkyGcO",
        "outputId": "d4d7cf92-a0ef-46e9-c621-918a175358f1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       MAL_ID                        show_name  \\\n",
              "0           1                     Cowboy Bebop   \n",
              "1           5           Cowboy Bebop:The Movie   \n",
              "2           6                           Trigun   \n",
              "3           7               Witch Hunter Robin   \n",
              "4           8           Beet the Vandel Buster   \n",
              "...       ...                              ...   \n",
              "17557   48481  Daomu Biji Zhi Qinling Shen Shu   \n",
              "17558   48483                     Mieruko-chan   \n",
              "17559   48488  Higurashi:When They Cry – SOTSU   \n",
              "17560   48491      Yama no Susume: Next Summit   \n",
              "17561   48492                    Scarlet Nexus   \n",
              "\n",
              "                                                  Genres  \n",
              "0        Action, Adventure, Comedy, Drama, Sci-Fi, Space  \n",
              "1                  Action, Drama, Mystery, Sci-Fi, Space  \n",
              "2      Action, Sci-Fi, Adventure, Comedy, Drama, Shounen  \n",
              "3      Action, Mystery, Police, Supernatural, Drama, ...  \n",
              "4              Adventure, Fantasy, Shounen, Supernatural  \n",
              "...                                                  ...  \n",
              "17557                   Adventure, Mystery, Supernatural  \n",
              "17558                       Comedy, Horror, Supernatural  \n",
              "17559  Mystery, Dementia, Horror, Psychological, Supe...  \n",
              "17560                   Adventure, Slice of Life, Comedy  \n",
              "17561                                    Action, Fantasy  \n",
              "\n",
              "[17562 rows x 3 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a03de90f-e5ba-45d7-9813-78b36ebd7fec\">\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>MAL_ID</th>\n",
              "      <th>show_name</th>\n",
              "      <th>Genres</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Cowboy Bebop</td>\n",
              "      <td>Action, Adventure, Comedy, Drama, Sci-Fi, Space</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5</td>\n",
              "      <td>Cowboy Bebop:The Movie</td>\n",
              "      <td>Action, Drama, Mystery, Sci-Fi, Space</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>6</td>\n",
              "      <td>Trigun</td>\n",
              "      <td>Action, Sci-Fi, Adventure, Comedy, Drama, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>7</td>\n",
              "      <td>Witch Hunter Robin</td>\n",
              "      <td>Action, Mystery, Police, Supernatural, Drama, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>8</td>\n",
              "      <td>Beet the Vandel Buster</td>\n",
              "      <td>Adventure, Fantasy, Shounen, Supernatural</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17557</th>\n",
              "      <td>48481</td>\n",
              "      <td>Daomu Biji Zhi Qinling Shen Shu</td>\n",
              "      <td>Adventure, Mystery, Supernatural</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17558</th>\n",
              "      <td>48483</td>\n",
              "      <td>Mieruko-chan</td>\n",
              "      <td>Comedy, Horror, Supernatural</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17559</th>\n",
              "      <td>48488</td>\n",
              "      <td>Higurashi:When They Cry – SOTSU</td>\n",
              "      <td>Mystery, Dementia, Horror, Psychological, Supe...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17560</th>\n",
              "      <td>48491</td>\n",
              "      <td>Yama no Susume: Next Summit</td>\n",
              "      <td>Adventure, Slice of Life, Comedy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17561</th>\n",
              "      <td>48492</td>\n",
              "      <td>Scarlet Nexus</td>\n",
              "      <td>Action, Fantasy</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>17562 rows × 3 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a03de90f-e5ba-45d7-9813-78b36ebd7fec')\"\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-a03de90f-e5ba-45d7-9813-78b36ebd7fec 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-a03de90f-e5ba-45d7-9813-78b36ebd7fec');\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": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(len(anime_list))\n",
        "print(len(synopsis['MAL_ID'].tolist()))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bV02J7ZUhIws",
        "outputId": "0c050a4a-e9a6-4178-912d-fa3d98dba328"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "16953\n",
            "17562\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There are some animes not in the synopsis."
      ],
      "metadata": {
        "id": "TtwoEC43p8ba"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "pd.set_option(\"max_colwidth\", None)\n",
        "\n",
        "# Find the similar animes.\n",
        "def find_similar_animes(anime, k=10):\n",
        "    try:\n",
        "      anime_id = int(synopsis[synopsis['show_name'] == anime].MAL_ID)\n",
        "      anime_id_encoded = anime_encoded[anime_id]                    # Encode the anime.\n",
        "      dist = np.dot(anime_weights, anime_weights[anime_id_encoded]) # Calculate the cosine similarity.\n",
        "      closest = np.argpartition(dist, -(k+1))[-(k+1):]              # Sort the score.\n",
        "\n",
        "      suggestion = []\n",
        "      \n",
        "      # Map to the anime name\n",
        "      for idx in closest:\n",
        "        score = dist[idx]\n",
        "        suggestion_idx = anime_decoded[idx]\n",
        "        row = synopsis[synopsis['MAL_ID'] == suggestion_idx]\n",
        "        anime_name = row['show_name'].values[0]\n",
        "        genre = row['Genres'].values[0]\n",
        "\n",
        "        suggestion.append({'name':anime_name,\n",
        "                          'score':score,\n",
        "                          'genre':genre})\n",
        "\n",
        "      suggestion_df = pd.DataFrame(suggestion).sort_values(by=\"score\", ascending=False)\n",
        "      return suggestion_df[suggestion_df['name']!= anime]\n",
        "\n",
        "    except:\n",
        "        print(f'{anime} is not found.')"
      ],
      "metadata": {
        "id": "O4ZNu_4FhkSk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "find_similar_animes('Dragon Ball Z')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "nN6lwBYNDkCD",
        "outputId": "719e8282-7114-41ad-8842-8b91a81e441d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                   name     score  \\\n",
              "10                                          Dragon Ball  0.888997   \n",
              "8                                             One Piece  0.752096   \n",
              "9                           Dragon Ball Z:Fusion Reborn  0.721909   \n",
              "6      Dragon Ball Z:Broly – The Legendary Super Saiyan  0.719839   \n",
              "4                                     Dragon Ball Z Kai  0.719088   \n",
              "5                             Yu Yu Hakusho:Ghost Files  0.702248   \n",
              "3         Dragon Ball Z Special 2:The History of Trunks  0.699071   \n",
              "2                  Dragon Ball Z Kai:The Final Chapters  0.682471   \n",
              "1   Dragon Ball Z Special 1:Bardock, The Father of Goku  0.655972   \n",
              "0                                     Dragon Ball Super  0.647954   \n",
              "\n",
              "                                                                     genre  \n",
              "10          Adventure, Comedy, Fantasy, Martial Arts, Shounen, Super Power  \n",
              "8          Action, Adventure, Comedy, Super Power, Drama, Fantasy, Shounen  \n",
              "9                      Action, Adventure, Comedy, Fantasy, Sci-Fi, Shounen  \n",
              "6                      Action, Adventure, Comedy, Fantasy, Sci-Fi, Shounen  \n",
              "4   Action, Adventure, Comedy, Fantasy, Martial Arts, Shounen, Super Power  \n",
              "5              Action, Comedy, Demons, Supernatural, Martial Arts, Shounen  \n",
              "3                                       Adventure, Drama, Fantasy, Shounen  \n",
              "2   Action, Adventure, Comedy, Super Power, Martial Arts, Fantasy, Shounen  \n",
              "1                              Adventure, Comedy, Fantasy, Sci-Fi, Shounen  \n",
              "0   Action, Adventure, Comedy, Super Power, Martial Arts, Fantasy, Shounen  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ee5bc1fa-20a4-4dd1-85dc-417e58ea2930\">\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>name</th>\n",
              "      <th>score</th>\n",
              "      <th>genre</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>Dragon Ball</td>\n",
              "      <td>0.888997</td>\n",
              "      <td>Adventure, Comedy, Fantasy, Martial Arts, Shounen, Super Power</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>One Piece</td>\n",
              "      <td>0.752096</td>\n",
              "      <td>Action, Adventure, Comedy, Super Power, Drama, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Dragon Ball Z:Fusion Reborn</td>\n",
              "      <td>0.721909</td>\n",
              "      <td>Action, Adventure, Comedy, Fantasy, Sci-Fi, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Dragon Ball Z:Broly – The Legendary Super Saiyan</td>\n",
              "      <td>0.719839</td>\n",
              "      <td>Action, Adventure, Comedy, Fantasy, Sci-Fi, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Dragon Ball Z Kai</td>\n",
              "      <td>0.719088</td>\n",
              "      <td>Action, Adventure, Comedy, Fantasy, Martial Arts, Shounen, Super Power</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Yu Yu Hakusho:Ghost Files</td>\n",
              "      <td>0.702248</td>\n",
              "      <td>Action, Comedy, Demons, Supernatural, Martial Arts, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Dragon Ball Z Special 2:The History of Trunks</td>\n",
              "      <td>0.699071</td>\n",
              "      <td>Adventure, Drama, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Dragon Ball Z Kai:The Final Chapters</td>\n",
              "      <td>0.682471</td>\n",
              "      <td>Action, Adventure, Comedy, Super Power, Martial Arts, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Dragon Ball Z Special 1:Bardock, The Father of Goku</td>\n",
              "      <td>0.655972</td>\n",
              "      <td>Adventure, Comedy, Fantasy, Sci-Fi, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Dragon Ball Super</td>\n",
              "      <td>0.647954</td>\n",
              "      <td>Action, Adventure, Comedy, Super Power, Martial Arts, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ee5bc1fa-20a4-4dd1-85dc-417e58ea2930')\"\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-ee5bc1fa-20a4-4dd1-85dc-417e58ea2930 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-ee5bc1fa-20a4-4dd1-85dc-417e58ea2930');\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": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def get_fav_animes(user):\n",
        "  fav_animes = df[df['user_id']==user]\n",
        "  rating90 = np.percentile(fav_animes['rating'], 90)\n",
        "  fav_animes = fav_animes[fav_animes['rating'] >= rating90]\n",
        "  fav_anime_id = fav_animes.sort_values(by='rating', ascending=False)['anime_id'].tolist()\n",
        "\n",
        "  fav_anime_df = synopsis[synopsis['MAL_ID'].isin(fav_anime_id)]\n",
        "  return fav_anime_df"
      ],
      "metadata": {
        "id": "pdbkV72E1ar9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "get_fav_animes(1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 519
        },
        "id": "TKrQ7ecA1ilp",
        "outputId": "0f3ed0af-8a99-4479-cc67-6c1b710430a3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       MAL_ID                                                    show_name  \\\n",
              "10         20                                                       Naruto   \n",
              "176       199                                                Spirited Away   \n",
              "1431     1575                          Code Geass:Lelouch of the Rebellion   \n",
              "1574     1735                                             Naruto:Shippuden   \n",
              "3971     5114                              Fullmetal Alchemist:Brotherhood   \n",
              "5683     9253                                                  Steins;Gate   \n",
              "6565    11577               Steins;Gate:The Movie − Load Region of Déjà Vu   \n",
              "7449    16498                                              Attack on Titan   \n",
              "9383    25777                                     Attack on Titan Season 2   \n",
              "10569   30484                                                Steins;Gate 0   \n",
              "11261   32188  Steins;Gate: Kyoukaimenjou no Missing Link - Divide By Zero   \n",
              "11308   32281                                                   Your Name.   \n",
              "13252   35760                                     Attack on Titan Season 3   \n",
              "14266   37450                   Rascal Does Not Dream of Bunny Girl Senpai   \n",
              "15926   40028                                 Attack on Titan Final Season   \n",
              "\n",
              "                                                                    Genres  \n",
              "10           Action, Adventure, Comedy, Super Power, Martial Arts, Shounen  \n",
              "176                                         Adventure, Supernatural, Drama  \n",
              "1431           Action, Military, Sci-Fi, Super Power, Drama, Mecha, School  \n",
              "1574         Action, Adventure, Comedy, Super Power, Martial Arts, Shounen  \n",
              "3971   Action, Military, Adventure, Comedy, Drama, Magic, Fantasy, Shounen  \n",
              "5683                                                      Thriller, Sci-Fi  \n",
              "6565                                                         Sci-Fi, Drama  \n",
              "7449       Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen  \n",
              "9383       Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen  \n",
              "10569                                                     Sci-Fi, Thriller  \n",
              "11261                                                     Sci-Fi, Thriller  \n",
              "11308                                 Romance, Supernatural, School, Drama  \n",
              "13252      Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen  \n",
              "14266                         Comedy, Supernatural, Drama, Romance, School  \n",
              "15926      Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0d1737df-975b-42e5-b0d8-b17779582626\">\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>MAL_ID</th>\n",
              "      <th>show_name</th>\n",
              "      <th>Genres</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>20</td>\n",
              "      <td>Naruto</td>\n",
              "      <td>Action, Adventure, Comedy, Super Power, Martial Arts, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>176</th>\n",
              "      <td>199</td>\n",
              "      <td>Spirited Away</td>\n",
              "      <td>Adventure, Supernatural, Drama</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1431</th>\n",
              "      <td>1575</td>\n",
              "      <td>Code Geass:Lelouch of the Rebellion</td>\n",
              "      <td>Action, Military, Sci-Fi, Super Power, Drama, Mecha, School</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1574</th>\n",
              "      <td>1735</td>\n",
              "      <td>Naruto:Shippuden</td>\n",
              "      <td>Action, Adventure, Comedy, Super Power, Martial Arts, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3971</th>\n",
              "      <td>5114</td>\n",
              "      <td>Fullmetal Alchemist:Brotherhood</td>\n",
              "      <td>Action, Military, Adventure, Comedy, Drama, Magic, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5683</th>\n",
              "      <td>9253</td>\n",
              "      <td>Steins;Gate</td>\n",
              "      <td>Thriller, Sci-Fi</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6565</th>\n",
              "      <td>11577</td>\n",
              "      <td>Steins;Gate:The Movie − Load Region of Déjà Vu</td>\n",
              "      <td>Sci-Fi, Drama</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7449</th>\n",
              "      <td>16498</td>\n",
              "      <td>Attack on Titan</td>\n",
              "      <td>Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9383</th>\n",
              "      <td>25777</td>\n",
              "      <td>Attack on Titan Season 2</td>\n",
              "      <td>Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10569</th>\n",
              "      <td>30484</td>\n",
              "      <td>Steins;Gate 0</td>\n",
              "      <td>Sci-Fi, Thriller</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11261</th>\n",
              "      <td>32188</td>\n",
              "      <td>Steins;Gate: Kyoukaimenjou no Missing Link - Divide By Zero</td>\n",
              "      <td>Sci-Fi, Thriller</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11308</th>\n",
              "      <td>32281</td>\n",
              "      <td>Your Name.</td>\n",
              "      <td>Romance, Supernatural, School, Drama</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13252</th>\n",
              "      <td>35760</td>\n",
              "      <td>Attack on Titan Season 3</td>\n",
              "      <td>Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14266</th>\n",
              "      <td>37450</td>\n",
              "      <td>Rascal Does Not Dream of Bunny Girl Senpai</td>\n",
              "      <td>Comedy, Supernatural, Drama, Romance, School</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15926</th>\n",
              "      <td>40028</td>\n",
              "      <td>Attack on Titan Final Season</td>\n",
              "      <td>Action, Military, Mystery, Super Power, Drama, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0d1737df-975b-42e5-b0d8-b17779582626')\"\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-0d1737df-975b-42e5-b0d8-b17779582626 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-0d1737df-975b-42e5-b0d8-b17779582626');\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": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def recommend_similar_animes(user, k=5):\n",
        "  fav_animes = get_fav_animes(user)['show_name'].values\n",
        "  fav_anime_set = set(fav_animes)\n",
        "  similar_animes = []\n",
        "  for anime in fav_animes:\n",
        "    try:\n",
        "      animes =  set(find_similar_animes(anime)['name'].values) - fav_anime_set\n",
        "      similar_animes += list(animes)\n",
        "    except:\n",
        "      continue\n",
        "\n",
        "  suggest_names = [*pd.Series(similar_animes).value_counts().head(k).index]\n",
        "  suggested_animes = synopsis[synopsis['show_name'].isin(suggest_names)]\n",
        "\n",
        "  return suggested_animes"
      ],
      "metadata": {
        "id": "ayBA3uVt1lsG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "recommend_similar_animes(1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "1OwizZCc8nhD",
        "outputId": "cf38830c-7f3d-44b7-ac3e-0f62d0035b87"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "      MAL_ID                               show_name  \\\n",
              "2656    2904  Code Geass:Lelouch of the Rebellion R2   \n",
              "3537    4181                   Clannad ~After Story~   \n",
              "7261   15417                       Gintama:Enchousen   \n",
              "8883   23273                       Your Lie in April   \n",
              "9886   28851                          A Silent Voice   \n",
              "\n",
              "                                                            Genres  \n",
              "2656           Action, Military, Sci-Fi, Super Power, Drama, Mecha  \n",
              "3537           Slice of Life, Comedy, Supernatural, Drama, Romance  \n",
              "7261  Action, Comedy, Historical, Parody, Samurai, Sci-Fi, Shounen  \n",
              "8883                        Drama, Music, Romance, School, Shounen  \n",
              "9886                                        Drama, School, Shounen  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7a472f5b-8665-4dcb-a52a-cc057be19461\">\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>MAL_ID</th>\n",
              "      <th>show_name</th>\n",
              "      <th>Genres</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2656</th>\n",
              "      <td>2904</td>\n",
              "      <td>Code Geass:Lelouch of the Rebellion R2</td>\n",
              "      <td>Action, Military, Sci-Fi, Super Power, Drama, Mecha</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3537</th>\n",
              "      <td>4181</td>\n",
              "      <td>Clannad ~After Story~</td>\n",
              "      <td>Slice of Life, Comedy, Supernatural, Drama, Romance</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7261</th>\n",
              "      <td>15417</td>\n",
              "      <td>Gintama:Enchousen</td>\n",
              "      <td>Action, Comedy, Historical, Parody, Samurai, Sci-Fi, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8883</th>\n",
              "      <td>23273</td>\n",
              "      <td>Your Lie in April</td>\n",
              "      <td>Drama, Music, Romance, School, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9886</th>\n",
              "      <td>28851</td>\n",
              "      <td>A Silent Voice</td>\n",
              "      <td>Drama, School, Shounen</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7a472f5b-8665-4dcb-a52a-cc057be19461')\"\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-7a472f5b-8665-4dcb-a52a-cc057be19461 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-7a472f5b-8665-4dcb-a52a-cc057be19461');\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": 68
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##4.2 Recommend animes which are watched by similar user."
      ],
      "metadata": {
        "id": "D07tSUQu1GQx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def find_similar_users(user, k=10):\n",
        "    try:\n",
        "      user_id_encoded = user_encoded[user]\n",
        "      dist = np.dot(user_weights, user_weights[user_id_encoded])\n",
        "      closest = np.argpartition(dist, -(k+1))[-(k+1):]\n",
        "\n",
        "      suggestion = []\n",
        "      for idx in closest:\n",
        "        score = dist[idx]\n",
        "        similar_user = user_decoded[idx]\n",
        "\n",
        "        suggestion.append({'similar_user':similar_user,\n",
        "                          'score':score})\n",
        "\n",
        "      suggestion_df = pd.DataFrame(suggestion).sort_values(by=\"score\", ascending=False)\n",
        "      return suggestion_df[suggestion_df['similar_user']!= user]\n",
        "\n",
        "    except:\n",
        "        print(f'{user} is not found.')"
      ],
      "metadata": {
        "id": "kYYKkfjrJoSF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "find_similar_users(32600)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "v2hrLWP9KY2z",
        "outputId": "1a6226af-64c5-428f-a6ae-812111d3b16a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    similar_user     score\n",
              "8          45156  0.526443\n",
              "9         141744  0.506963\n",
              "6            117  0.491387\n",
              "4          12454  0.487128\n",
              "7          84069  0.472681\n",
              "5         120314  0.469694\n",
              "10        138292  0.453647\n",
              "2         101731  0.451429\n",
              "1         110124  0.448902\n",
              "0          59194  0.440300"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6e602f94-ba52-4fa3-a5eb-3057975bfd61\">\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>similar_user</th>\n",
              "      <th>score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>45156</td>\n",
              "      <td>0.526443</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>141744</td>\n",
              "      <td>0.506963</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>117</td>\n",
              "      <td>0.491387</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>12454</td>\n",
              "      <td>0.487128</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>84069</td>\n",
              "      <td>0.472681</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>120314</td>\n",
              "      <td>0.469694</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>138292</td>\n",
              "      <td>0.453647</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>101731</td>\n",
              "      <td>0.451429</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>110124</td>\n",
              "      <td>0.448902</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>59194</td>\n",
              "      <td>0.440300</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6e602f94-ba52-4fa3-a5eb-3057975bfd61')\"\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-6e602f94-ba52-4fa3-a5eb-3057975bfd61 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-6e602f94-ba52-4fa3-a5eb-3057975bfd61');\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": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def recommend_animes_by_user(user, k=5):\n",
        "  fav_animes = get_fav_animes(user)['show_name'].values\n",
        "  fav_anime_set = set(fav_animes)\n",
        "\n",
        "  similar_users = find_similar_users(user)['similar_user'].values\n",
        "  anime_candidates = []\n",
        "  for similar_user in similar_users:\n",
        "    try:\n",
        "      fav_animes_of_similar_user = get_fav_animes(similar_user)['show_name'].values\n",
        "      anime_candidates += list(set(fav_animes_of_similar_user) - fav_anime_set)\n",
        "    except:\n",
        "      continue\n",
        "  \n",
        "  suggest_names = [*pd.Series(anime_candidates).value_counts().head(k).index]\n",
        "  suggested_animes = synopsis[synopsis['show_name'].isin(suggest_names)]\n",
        "\n",
        "  return suggested_animes\n",
        "  "
      ],
      "metadata": {
        "id": "zRwACjZXy0eA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "recommend_animes_by_user(1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "TtyJxoNQ3Mor",
        "outputId": "849e3271-3e2b-45d7-9c30-dda7a0cb9f11"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       MAL_ID                        show_name  \\\n",
              "115       136                  Hunter x Hunter   \n",
              "6474    11061                  Hunter x Hunter   \n",
              "8883    23273                Your Lie in April   \n",
              "14306   37510                Mob Psycho 100 II   \n",
              "14485   37779           The Promised Neverland   \n",
              "14963   38524  Attack on Titan Season 3 Part 2   \n",
              "\n",
              "                                                                Genres  \n",
              "115                   Action, Adventure, Super Power, Fantasy, Shounen  \n",
              "6474                  Action, Adventure, Fantasy, Shounen, Super Power  \n",
              "8883                            Drama, Music, Romance, School, Shounen  \n",
              "14306                      Action, Slice of Life, Comedy, Supernatural  \n",
              "14485        Sci-Fi, Mystery, Horror, Psychological, Thriller, Shounen  \n",
              "14963  Action, Drama, Fantasy, Military, Mystery, Shounen, Super Power  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0e0757b9-457f-4fb9-877a-c3a39cea8008\">\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>MAL_ID</th>\n",
              "      <th>show_name</th>\n",
              "      <th>Genres</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>115</th>\n",
              "      <td>136</td>\n",
              "      <td>Hunter x Hunter</td>\n",
              "      <td>Action, Adventure, Super Power, Fantasy, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6474</th>\n",
              "      <td>11061</td>\n",
              "      <td>Hunter x Hunter</td>\n",
              "      <td>Action, Adventure, Fantasy, Shounen, Super Power</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8883</th>\n",
              "      <td>23273</td>\n",
              "      <td>Your Lie in April</td>\n",
              "      <td>Drama, Music, Romance, School, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14306</th>\n",
              "      <td>37510</td>\n",
              "      <td>Mob Psycho 100 II</td>\n",
              "      <td>Action, Slice of Life, Comedy, Supernatural</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14485</th>\n",
              "      <td>37779</td>\n",
              "      <td>The Promised Neverland</td>\n",
              "      <td>Sci-Fi, Mystery, Horror, Psychological, Thriller, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14963</th>\n",
              "      <td>38524</td>\n",
              "      <td>Attack on Titan Season 3 Part 2</td>\n",
              "      <td>Action, Drama, Fantasy, Military, Mystery, Shounen, Super Power</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0e0757b9-457f-4fb9-877a-c3a39cea8008')\"\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-0e0757b9-457f-4fb9-877a-c3a39cea8008 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-0e0757b9-457f-4fb9-877a-c3a39cea8008');\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": 69
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 4.3 Recommend top rating animes"
      ],
      "metadata": {
        "id": "wyvcnbMgApyR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def recommend_top_animes(user, k=5):\n",
        "  watched_anime_df = df[df.user_id==user]\n",
        "  unwatched_anime_df = synopsis[~synopsis[\"MAL_ID\"].isin(watched_anime_df['anime_id'].values)]\n",
        "  unwatched_animes = list(set(unwatched_anime_df['MAL_ID']).intersection(set(anime_encoded.keys())))\n",
        "  unwatched_animes = [[anime_encoded.get(x)] for x in unwatched_animes]\n",
        "  \n",
        "  user_id = user_encoded.get(user)\n",
        "  X_array = [ np.array([user_id]*len(unwatched_animes)), np.array(unwatched_animes) ]\n",
        "  ratings = model.predict(X_array).flatten()\n",
        "  top_ratings_indices = (-ratings).argsort()[:k]\n",
        "  recommended_anime_ids = [anime_decoded.get(unwatched_animes[x][0]) for x in top_ratings_indices]\n",
        "\n",
        "  suggestion = []\n",
        "  top_rated_ids = []\n",
        "  for idx, anime_id in enumerate(unwatched_animes):\n",
        "    rating = ratings[idx]\n",
        "    id_ = anime_decoded.get(anime_id[0])\n",
        "    \n",
        "    if id_ in recommended_anime_ids:\n",
        "        top_rated_ids.append(id_)\n",
        "        try:\n",
        "            row = synopsis[synopsis['MAL_ID'] == id_]\n",
        "            anime_name = row['show_name'].values[0]\n",
        "            genre = row['Genres'].values[0]\n",
        "            suggestion.append({'name':anime_name,\n",
        "                              'pred_rating':rating,\n",
        "                              'genre':genre})\n",
        "        except:\n",
        "            continue\n",
        "            \n",
        "  suggestion_df = pd.DataFrame(suggestion).sort_values(by=\"pred_rating\", ascending=False)\n",
        "  return suggestion_df"
      ],
      "metadata": {
        "id": "C5dy7OvLAvZa"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "recommend_top_animes(1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eEMIk-Jb1nnp",
        "outputId": "468d59e9-e6ab-4018-97e4-b7adf8fdb72f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                                                                  name  \\\n",
              "1                                                                                   Gintama: The Final   \n",
              "3  Maou Gakuin no Futekigousha: Shijou Saikyou no Maou no Shiso, Tensei shite Shison-tachi no Gakkou e   \n",
              "4                                                                                     Gintama Season 4   \n",
              "2  Maou Gakuin no Futekigousha: Shijou Saikyou no Maou no Shiso, Tensei shite Shison-tachi no Gakkou e   \n",
              "0                                                                       Sensou: Kodomo-tachi no Yuigon   \n",
              "\n",
              "   pred_rating  \\\n",
              "1     0.980847   \n",
              "3     0.965877   \n",
              "4     0.957808   \n",
              "2     0.955009   \n",
              "0     0.952838   \n",
              "\n",
              "                                                                 genre  \n",
              "1  Action, Sci-Fi, Comedy, Historical, Parody, Drama, Samurai, Shounen  \n",
              "3                               Action, Demons, Magic, Fantasy, School  \n",
              "4         Action, Comedy, Historical, Parody, Samurai, Sci-Fi, Shounen  \n",
              "2                                               Magic, Fantasy, School  \n",
              "0                                                    Drama, Historical  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f28a08c4-b0a9-4b71-8502-a211650cb4cf\">\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>name</th>\n",
              "      <th>pred_rating</th>\n",
              "      <th>genre</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Gintama: The Final</td>\n",
              "      <td>0.980847</td>\n",
              "      <td>Action, Sci-Fi, Comedy, Historical, Parody, Drama, Samurai, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Maou Gakuin no Futekigousha: Shijou Saikyou no Maou no Shiso, Tensei shite Shison-tachi no Gakkou e</td>\n",
              "      <td>0.965877</td>\n",
              "      <td>Action, Demons, Magic, Fantasy, School</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Gintama Season 4</td>\n",
              "      <td>0.957808</td>\n",
              "      <td>Action, Comedy, Historical, Parody, Samurai, Sci-Fi, Shounen</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Maou Gakuin no Futekigousha: Shijou Saikyou no Maou no Shiso, Tensei shite Shison-tachi no Gakkou e</td>\n",
              "      <td>0.955009</td>\n",
              "      <td>Magic, Fantasy, School</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Sensou: Kodomo-tachi no Yuigon</td>\n",
              "      <td>0.952838</td>\n",
              "      <td>Drama, Historical</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f28a08c4-b0a9-4b71-8502-a211650cb4cf')\"\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-f28a08c4-b0a9-4b71-8502-a211650cb4cf 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-f28a08c4-b0a9-4b71-8502-a211650cb4cf');\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": 70
        }
      ]
    }
  ]
}