content-based-and-tensorflow-recommender-system.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.7.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":1187,"sourceType":"datasetVersion","datasetId":626},{"sourceId":4508,"sourceType":"datasetVersion","datasetId":138},{"sourceId":4966,"sourceType":"datasetVersion","datasetId":2559},{"sourceId":6663,"sourceType":"datasetVersion","datasetId":3405},{"sourceId":10279,"sourceType":"datasetVersion","datasetId":7181},{"sourceId":16363,"sourceType":"datasetVersion","datasetId":11616},{"sourceId":77759,"sourceType":"datasetVersion","datasetId":339},{"sourceId":118283,"sourceType":"datasetVersion","datasetId":60876},{"sourceId":127736,"sourceType":"datasetVersion","datasetId":64890},{"sourceId":320111,"sourceType":"datasetVersion","datasetId":134715},{"sourceId":792972,"sourceType":"datasetVersion","datasetId":1636},{"sourceId":1464660,"sourceType":"datasetVersion","datasetId":858948},{"sourceId":1863460,"sourceType":"datasetVersion","datasetId":1109013},{"sourceId":1898721,"sourceType":"datasetVersion","datasetId":1131493},{"sourceId":2456241,"sourceType":"datasetVersion","datasetId":2745},{"sourceId":2654038,"sourceType":"datasetVersion","datasetId":434238},{"sourceId":2698943,"sourceType":"datasetVersion","datasetId":1643699},{"sourceId":2864236,"sourceType":"datasetVersion","datasetId":1624191},{"sourceId":2935141,"sourceType":"datasetVersion","datasetId":669193},{"sourceId":3247929,"sourceType":"datasetVersion","datasetId":1968448},{"sourceId":3989707,"sourceType":"datasetVersion","datasetId":2178661},{"sourceId":7398613,"sourceType":"datasetVersion","datasetId":592158}],"dockerImageVersionId":30213,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"source":"<a href=\"https://www.kaggle.com/code/faiqueali/tensorflow-movie-recommender-system?scriptVersionId=159742868\" target=\"_blank\"><img align=\"left\" alt=\"Kaggle\" title=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"></a>","metadata":{},"cell_type":"markdown"},{"cell_type":"markdown","source":"<h1 style=\"font-family: Trebuchet MS; padding: 12px; font-size: 30px; color: #081c15; text-align: center; line-height: 1.25;\">Recommender System<br><span style=\"color: #1b4332; font-size: 48px\"><b>🎥Movies Recommendation📹</b></span><br><span style=\"color: #52b788; font-size: 20px\">📀Using Deep Learning</span></h1>\n<hr>","metadata":{"papermill":{"duration":0.017833,"end_time":"2022-08-08T09:10:20.410172","exception":false,"start_time":"2022-08-08T09:10:20.392339","status":"completed"},"tags":[]}},{"cell_type":"code","source":"!pip install -q tensorflow-recommenders","metadata":{"_kg_hide-input":true,"_kg_hide-output":true,"papermill":{"duration":92.090907,"end_time":"2022-08-08T09:11:52.708649","exception":false,"start_time":"2022-08-08T09:10:20.617742","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:20:07.960531Z","iopub.execute_input":"2024-01-20T13:20:07.96145Z","iopub.status.idle":"2024-01-20T13:21:28.847649Z","shell.execute_reply.started":"2024-01-20T13:20:07.961345Z","shell.execute_reply":"2024-01-20T13:21:28.846094Z"},"trusted":true},"execution_count":1,"outputs":[{"name":"stdout","text":"\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\ntfx-bsl 1.9.0 requires pyarrow<6,>=1, but you have pyarrow 8.0.0 which is incompatible.\ntensorflow-transform 1.9.0 requires pyarrow<6,>=1, but you have pyarrow 8.0.0 which is incompatible.\ntensorflow-transform 1.9.0 requires tensorflow!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,<2.10,>=1.15.5, but you have tensorflow 2.11.0 which is incompatible.\ntensorflow-io 0.21.0 requires tensorflow<2.7.0,>=2.6.0, but you have tensorflow 2.11.0 which is incompatible.\ntensorflow-io 0.21.0 requires tensorflow-io-gcs-filesystem==0.21.0, but you have tensorflow-io-gcs-filesystem 0.34.0 which is incompatible.\njaxlib 0.3.14 requires flatbuffers<3.0,>=1.12, but you have flatbuffers 23.5.26 which is incompatible.\u001b[0m\u001b[31m\n\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n\u001b[0m","output_type":"stream"}]},{"cell_type":"markdown","source":"# Imports","metadata":{}},{"cell_type":"code","source":"import string\nimport re\nimport pandas as pd\nimport numpy as np\nimport seaborn as sns\nimport matplotlib.pyplot as plt\nimport tensorflow as tf\nimport tensorflow_recommenders as tfrs\nfrom collections import Counter\nfrom typing import Dict, Text\nfrom ast import literal_eval\nfrom datetime import datetime\nfrom wordcloud import WordCloud\nfrom sklearn.preprocessing import MinMaxScaler\nfrom sklearn.feature_extraction.text import TfidfVectorizer\nfrom sklearn.metrics.pairwise import cosine_similarity\n\nimport warnings\nwarnings.filterwarnings('ignore')","metadata":{"_kg_hide-input":true,"_kg_hide-output":true,"papermill":{"duration":4.977321,"end_time":"2022-08-08T09:11:57.701696","exception":false,"start_time":"2022-08-08T09:11:52.724375","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:28.850721Z","iopub.execute_input":"2024-01-20T13:21:28.851126Z","iopub.status.idle":"2024-01-20T13:21:34.436821Z","shell.execute_reply.started":"2024-01-20T13:21:28.851075Z","shell.execute_reply":"2024-01-20T13:21:34.435008Z"},"trusted":true},"execution_count":2,"outputs":[]},{"cell_type":"markdown","source":"# Code","metadata":{}},{"cell_type":"code","source":"credits = pd.read_csv('../input/the-movies-dataset/credits.csv')\nkeywords = pd.read_csv('../input/the-movies-dataset/keywords.csv')\nmovies = pd.read_csv('../input/the-movies-dataset/movies_metadata.csv').\\\n                     drop(['belongs_to_collection', 'homepage', 'imdb_id', 'poster_path', 'status', 'title', 'video'], axis=1).\\\n                     drop([19730, 29503, 35587]) # Incorrect data type\n\nmovies['id'] = movies['id'].astype('int64')\n\ndf = movies.merge(keywords, on='id').\\\n    merge(credits, on='id')\n\ndf['original_language'] = df['original_language'].fillna('')\ndf['runtime'] = df['runtime'].fillna(0)\ndf['tagline'] = df['tagline'].fillna('')\n\ndf.dropna(inplace=True)","metadata":{"_kg_hide-input":true,"_kg_hide-output":false,"papermill":{"duration":5.890113,"end_time":"2022-08-08T09:12:03.608246","exception":false,"start_time":"2022-08-08T09:11:57.718133","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:34.438786Z","iopub.execute_input":"2024-01-20T13:21:34.439757Z","iopub.status.idle":"2024-01-20T13:21:40.334793Z","shell.execute_reply.started":"2024-01-20T13:21:34.439707Z","shell.execute_reply":"2024-01-20T13:21:40.333505Z"},"trusted":true},"execution_count":3,"outputs":[]},{"cell_type":"code","source":"df.head()","metadata":{"papermill":{"duration":0.054458,"end_time":"2022-08-08T09:13:04.261672","exception":false,"start_time":"2022-08-08T09:13:04.207214","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:40.338326Z","iopub.execute_input":"2024-01-20T13:21:40.338831Z","iopub.status.idle":"2024-01-20T13:21:40.378601Z","shell.execute_reply.started":"2024-01-20T13:21:40.338784Z","shell.execute_reply":"2024-01-20T13:21:40.377397Z"},"trusted":true},"execution_count":4,"outputs":[{"execution_count":4,"output_type":"execute_result","data":{"text/plain":"   adult    budget                                             genres     id  \\\n0  False  30000000  [{'id': 16, 'name': 'Animation'}, {'id': 35, '...    862   \n1  False  65000000  [{'id': 12, 'name': 'Adventure'}, {'id': 14, '...   8844   \n2  False         0  [{'id': 10749, 'name': 'Romance'}, {'id': 35, ...  15602   \n3  False  16000000  [{'id': 35, 'name': 'Comedy'}, {'id': 18, 'nam...  31357   \n4  False         0                     [{'id': 35, 'name': 'Comedy'}]  11862   \n\n  original_language               original_title  \\\n0                en                    Toy Story   \n1                en                      Jumanji   \n2                en             Grumpier Old Men   \n3                en            Waiting to Exhale   \n4                en  Father of the Bride Part II   \n\n                                            overview popularity  \\\n0  Led by Woody, Andy's toys live happily in his ...  21.946943   \n1  When siblings Judy and Peter discover an encha...  17.015539   \n2  A family wedding reignites the ancient feud be...    11.7129   \n3  Cheated on, mistreated and stepped on, the wom...   3.859495   \n4  Just when George Banks has recovered from his ...   8.387519   \n\n                                production_companies  \\\n0     [{'name': 'Pixar Animation Studios', 'id': 3}]   \n1  [{'name': 'TriStar Pictures', 'id': 559}, {'na...   \n2  [{'name': 'Warner Bros.', 'id': 6194}, {'name'...   \n3  [{'name': 'Twentieth Century Fox Film Corporat...   \n4  [{'name': 'Sandollar Productions', 'id': 5842}...   \n\n                                production_countries release_date  \\\n0  [{'iso_3166_1': 'US', 'name': 'United States o...   1995-10-30   \n1  [{'iso_3166_1': 'US', 'name': 'United States o...   1995-12-15   \n2  [{'iso_3166_1': 'US', 'name': 'United States o...   1995-12-22   \n3  [{'iso_3166_1': 'US', 'name': 'United States o...   1995-12-22   \n4  [{'iso_3166_1': 'US', 'name': 'United States o...   1995-02-10   \n\n       revenue  runtime                                   spoken_languages  \\\n0  373554033.0     81.0           [{'iso_639_1': 'en', 'name': 'English'}]   \n1  262797249.0    104.0  [{'iso_639_1': 'en', 'name': 'English'}, {'iso...   \n2          0.0    101.0           [{'iso_639_1': 'en', 'name': 'English'}]   \n3   81452156.0    127.0           [{'iso_639_1': 'en', 'name': 'English'}]   \n4   76578911.0    106.0           [{'iso_639_1': 'en', 'name': 'English'}]   \n\n                                             tagline  vote_average  \\\n0                                                              7.7   \n1          Roll the dice and unleash the excitement!           6.9   \n2  Still Yelling. Still Fighting. Still Ready for...           6.5   \n3  Friends are the people who let you be yourself...           6.1   \n4  Just When His World Is Back To Normal... He's ...           5.7   \n\n   vote_count                                           keywords  \\\n0      5415.0  [{'id': 931, 'name': 'jealousy'}, {'id': 4290,...   \n1      2413.0  [{'id': 10090, 'name': 'board game'}, {'id': 1...   \n2        92.0  [{'id': 1495, 'name': 'fishing'}, {'id': 12392...   \n3        34.0  [{'id': 818, 'name': 'based on novel'}, {'id':...   \n4       173.0  [{'id': 1009, 'name': 'baby'}, {'id': 1599, 'n...   \n\n                                                cast  \\\n0  [{'cast_id': 14, 'character': 'Woody (voice)',...   \n1  [{'cast_id': 1, 'character': 'Alan Parrish', '...   \n2  [{'cast_id': 2, 'character': 'Max Goldman', 'c...   \n3  [{'cast_id': 1, 'character': \"Savannah 'Vannah...   \n4  [{'cast_id': 1, 'character': 'George Banks', '...   \n\n                                                crew  \n0  [{'credit_id': '52fe4284c3a36847f8024f49', 'de...  \n1  [{'credit_id': '52fe44bfc3a36847f80a7cd1', 'de...  \n2  [{'credit_id': '52fe466a9251416c75077a89', 'de...  \n3  [{'credit_id': '52fe44779251416c91011acb', 'de...  \n4  [{'credit_id': '52fe44959251416c75039ed7', 'de...  ","text/html":"<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>adult</th>\n      <th>budget</th>\n      <th>genres</th>\n      <th>id</th>\n      <th>original_language</th>\n      <th>original_title</th>\n      <th>overview</th>\n      <th>popularity</th>\n      <th>production_companies</th>\n      <th>production_countries</th>\n      <th>release_date</th>\n      <th>revenue</th>\n      <th>runtime</th>\n      <th>spoken_languages</th>\n      <th>tagline</th>\n      <th>vote_average</th>\n      <th>vote_count</th>\n      <th>keywords</th>\n      <th>cast</th>\n      <th>crew</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>False</td>\n      <td>30000000</td>\n      <td>[{'id': 16, 'name': 'Animation'}, {'id': 35, '...</td>\n      <td>862</td>\n      <td>en</td>\n      <td>Toy Story</td>\n      <td>Led by Woody, Andy's toys live happily in his ...</td>\n      <td>21.946943</td>\n      <td>[{'name': 'Pixar Animation Studios', 'id': 3}]</td>\n      <td>[{'iso_3166_1': 'US', 'name': 'United States o...</td>\n      <td>1995-10-30</td>\n      <td>373554033.0</td>\n      <td>81.0</td>\n      <td>[{'iso_639_1': 'en', 'name': 'English'}]</td>\n      <td></td>\n      <td>7.7</td>\n      <td>5415.0</td>\n      <td>[{'id': 931, 'name': 'jealousy'}, {'id': 4290,...</td>\n      <td>[{'cast_id': 14, 'character': 'Woody (voice)',...</td>\n      <td>[{'credit_id': '52fe4284c3a36847f8024f49', 'de...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>False</td>\n      <td>65000000</td>\n      <td>[{'id': 12, 'name': 'Adventure'}, {'id': 14, '...</td>\n      <td>8844</td>\n      <td>en</td>\n      <td>Jumanji</td>\n      <td>When siblings Judy and Peter discover an encha...</td>\n      <td>17.015539</td>\n      <td>[{'name': 'TriStar Pictures', 'id': 559}, {'na...</td>\n      <td>[{'iso_3166_1': 'US', 'name': 'United States o...</td>\n      <td>1995-12-15</td>\n      <td>262797249.0</td>\n      <td>104.0</td>\n      <td>[{'iso_639_1': 'en', 'name': 'English'}, {'iso...</td>\n      <td>Roll the dice and unleash the excitement!</td>\n      <td>6.9</td>\n      <td>2413.0</td>\n      <td>[{'id': 10090, 'name': 'board game'}, {'id': 1...</td>\n      <td>[{'cast_id': 1, 'character': 'Alan Parrish', '...</td>\n      <td>[{'credit_id': '52fe44bfc3a36847f80a7cd1', 'de...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>False</td>\n      <td>0</td>\n      <td>[{'id': 10749, 'name': 'Romance'}, {'id': 35, ...</td>\n      <td>15602</td>\n      <td>en</td>\n      <td>Grumpier Old Men</td>\n      <td>A family wedding reignites the ancient feud be...</td>\n      <td>11.7129</td>\n      <td>[{'name': 'Warner Bros.', 'id': 6194}, {'name'...</td>\n      <td>[{'iso_3166_1': 'US', 'name': 'United States o...</td>\n      <td>1995-12-22</td>\n      <td>0.0</td>\n      <td>101.0</td>\n      <td>[{'iso_639_1': 'en', 'name': 'English'}]</td>\n      <td>Still Yelling. Still Fighting. Still Ready for...</td>\n      <td>6.5</td>\n      <td>92.0</td>\n      <td>[{'id': 1495, 'name': 'fishing'}, {'id': 12392...</td>\n      <td>[{'cast_id': 2, 'character': 'Max Goldman', 'c...</td>\n      <td>[{'credit_id': '52fe466a9251416c75077a89', 'de...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>False</td>\n      <td>16000000</td>\n      <td>[{'id': 35, 'name': 'Comedy'}, {'id': 18, 'nam...</td>\n      <td>31357</td>\n      <td>en</td>\n      <td>Waiting to Exhale</td>\n      <td>Cheated on, mistreated and stepped on, the wom...</td>\n      <td>3.859495</td>\n      <td>[{'name': 'Twentieth Century Fox Film Corporat...</td>\n      <td>[{'iso_3166_1': 'US', 'name': 'United States o...</td>\n      <td>1995-12-22</td>\n      <td>81452156.0</td>\n      <td>127.0</td>\n      <td>[{'iso_639_1': 'en', 'name': 'English'}]</td>\n      <td>Friends are the people who let you be yourself...</td>\n      <td>6.1</td>\n      <td>34.0</td>\n      <td>[{'id': 818, 'name': 'based on novel'}, {'id':...</td>\n      <td>[{'cast_id': 1, 'character': \"Savannah 'Vannah...</td>\n      <td>[{'credit_id': '52fe44779251416c91011acb', 'de...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>False</td>\n      <td>0</td>\n      <td>[{'id': 35, 'name': 'Comedy'}]</td>\n      <td>11862</td>\n      <td>en</td>\n      <td>Father of the Bride Part II</td>\n      <td>Just when George Banks has recovered from his ...</td>\n      <td>8.387519</td>\n      <td>[{'name': 'Sandollar Productions', 'id': 5842}...</td>\n      <td>[{'iso_3166_1': 'US', 'name': 'United States o...</td>\n      <td>1995-02-10</td>\n      <td>76578911.0</td>\n      <td>106.0</td>\n      <td>[{'iso_639_1': 'en', 'name': 'English'}]</td>\n      <td>Just When His World Is Back To Normal... He's ...</td>\n      <td>5.7</td>\n      <td>173.0</td>\n      <td>[{'id': 1009, 'name': 'baby'}, {'id': 1599, 'n...</td>\n      <td>[{'cast_id': 1, 'character': 'George Banks', '...</td>\n      <td>[{'credit_id': '52fe44959251416c75039ed7', 'de...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"df.info()","metadata":{"_kg_hide-input":true,"_kg_hide-output":true,"papermill":{"duration":0.09014,"end_time":"2022-08-08T09:13:04.368549","exception":false,"start_time":"2022-08-08T09:13:04.278409","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:40.380221Z","iopub.execute_input":"2024-01-20T13:21:40.381014Z","iopub.status.idle":"2024-01-20T13:21:40.489534Z","shell.execute_reply.started":"2024-01-20T13:21:40.380964Z","shell.execute_reply":"2024-01-20T13:21:40.487985Z"},"trusted":true},"execution_count":5,"outputs":[{"name":"stdout","text":"<class 'pandas.core.frame.DataFrame'>\nInt64Index: 45558 entries, 0 to 46627\nData columns (total 20 columns):\n #   Column                Non-Null Count  Dtype  \n---  ------                --------------  -----  \n 0   adult                 45558 non-null  object \n 1   budget                45558 non-null  object \n 2   genres                45558 non-null  object \n 3   id                    45558 non-null  int64  \n 4   original_language     45558 non-null  object \n 5   original_title        45558 non-null  object \n 6   overview              45558 non-null  object \n 7   popularity            45558 non-null  object \n 8   production_companies  45558 non-null  object \n 9   production_countries  45558 non-null  object \n 10  release_date          45558 non-null  object \n 11  revenue               45558 non-null  float64\n 12  runtime               45558 non-null  float64\n 13  spoken_languages      45558 non-null  object \n 14  tagline               45558 non-null  object \n 15  vote_average          45558 non-null  float64\n 16  vote_count            45558 non-null  float64\n 17  keywords              45558 non-null  object \n 18  cast                  45558 non-null  object \n 19  crew                  45558 non-null  object \ndtypes: float64(4), int64(1), object(15)\nmemory usage: 7.3+ MB\n","output_type":"stream"}]},{"cell_type":"code","source":"df['release_date'] = pd.to_datetime(df['release_date'])\ndf['budget'] = df['budget'].astype('float64')\ndf['popularity'] = df['popularity'].astype('float64')","metadata":{"_kg_hide-input":true,"_kg_hide-output":false,"papermill":{"duration":0.088597,"end_time":"2022-08-08T09:13:04.473926","exception":false,"start_time":"2022-08-08T09:13:04.385329","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:40.491246Z","iopub.execute_input":"2024-01-20T13:21:40.491624Z","iopub.status.idle":"2024-01-20T13:21:40.579185Z","shell.execute_reply.started":"2024-01-20T13:21:40.491587Z","shell.execute_reply":"2024-01-20T13:21:40.577676Z"},"trusted":true},"execution_count":6,"outputs":[]},{"cell_type":"markdown","source":"<a id=2 a/>\n<p style=\"background-color:#368f8b;font-family:Trebuchet MS;font-weight:bold;color:#eff7f6;font-size:40px;text-align:center;border-radius:100px 100px\">Recommender System</p>","metadata":{"papermill":{"duration":0.043553,"end_time":"2022-08-08T09:13:34.242908","exception":false,"start_time":"2022-08-08T09:13:34.199355","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"<a id=4 a/>\n<h1 style=\"font-family: Trebuchet MS; font-size: 25px; color: #3a5a40; text-align: left; \"><b>Using Deep Learning</b></h1>\n\nOfficial documentation: https://www.tensorflow.org/recommenders\n\nTensorflow comes with a library called TensorFlow Recommenders (TFRS) for building a recommender system. It's built on Keras and aims to have a gentle learning curve while still giving you the flexibility to build complex models.","metadata":{"papermill":{"duration":0.044313,"end_time":"2022-08-08T09:13:57.074562","exception":false,"start_time":"2022-08-08T09:13:57.030249","status":"completed"},"tags":[]}},{"cell_type":"code","source":"ratings_df = pd.read_csv('../input/the-movies-dataset/ratings_small.csv')\n\nratings_df['date'] = ratings_df['timestamp'].apply(lambda x: datetime.fromtimestamp(x))\nratings_df.drop('timestamp', axis=1, inplace=True)\n\nratings_df = ratings_df.merge(df[['id', 'original_title', 'genres', 'overview']], left_on='movieId',right_on='id', how='left')\nratings_df = ratings_df[~ratings_df['id'].isna()]\nratings_df.drop('id', axis=1, inplace=True)\nratings_df.reset_index(drop=True, inplace=True)\n\nratings_df.head()","metadata":{"papermill":{"duration":0.254468,"end_time":"2022-08-08T09:13:57.373234","exception":false,"start_time":"2022-08-08T09:13:57.118766","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:40.581021Z","iopub.execute_input":"2024-01-20T13:21:40.581407Z","iopub.status.idle":"2024-01-20T13:21:40.869768Z","shell.execute_reply.started":"2024-01-20T13:21:40.581374Z","shell.execute_reply":"2024-01-20T13:21:40.868527Z"},"trusted":true},"execution_count":7,"outputs":[{"execution_count":7,"output_type":"execute_result","data":{"text/plain":"   userId  movieId  rating                date  \\\n0       1     1371     2.5 2009-12-14 02:52:15   \n1       1     1405     1.0 2009-12-14 02:53:23   \n2       1     2105     4.0 2009-12-14 02:52:19   \n3       1     2193     2.0 2009-12-14 02:53:18   \n4       1     2294     2.0 2009-12-14 02:51:48   \n\n                   original_title  \\\n0                       Rocky III   \n1                           Greed   \n2                    American Pie   \n3                        My Tutor   \n4  Jay and Silent Bob Strike Back   \n\n                                              genres  \\\n0                      [{'id': 18, 'name': 'Drama'}]   \n1  [{'id': 18, 'name': 'Drama'}, {'id': 36, 'name...   \n2  [{'id': 35, 'name': 'Comedy'}, {'id': 10749, '...   \n3  [{'id': 35, 'name': 'Comedy'}, {'id': 18, 'nam...   \n4                     [{'id': 35, 'name': 'Comedy'}]   \n\n                                            overview  \n0  Now the world champion, Rocky Balboa is living...  \n1  Greed is the classic 1924 silent film by Erich...  \n2  At a high-school party, four friends find that...  \n3  High school senior Bobby Chrystal fails his Fr...  \n4  When Jay and Silent Bob learn that their comic...  ","text/html":"<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>userId</th>\n      <th>movieId</th>\n      <th>rating</th>\n      <th>date</th>\n      <th>original_title</th>\n      <th>genres</th>\n      <th>overview</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1</td>\n      <td>1371</td>\n      <td>2.5</td>\n      <td>2009-12-14 02:52:15</td>\n      <td>Rocky III</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>Now the world champion, Rocky Balboa is living...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>1405</td>\n      <td>1.0</td>\n      <td>2009-12-14 02:53:23</td>\n      <td>Greed</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 36, 'name...</td>\n      <td>Greed is the classic 1924 silent film by Erich...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>1</td>\n      <td>2105</td>\n      <td>4.0</td>\n      <td>2009-12-14 02:52:19</td>\n      <td>American Pie</td>\n      <td>[{'id': 35, 'name': 'Comedy'}, {'id': 10749, '...</td>\n      <td>At a high-school party, four friends find that...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>1</td>\n      <td>2193</td>\n      <td>2.0</td>\n      <td>2009-12-14 02:53:18</td>\n      <td>My Tutor</td>\n      <td>[{'id': 35, 'name': 'Comedy'}, {'id': 18, 'nam...</td>\n      <td>High school senior Bobby Chrystal fails his Fr...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>1</td>\n      <td>2294</td>\n      <td>2.0</td>\n      <td>2009-12-14 02:51:48</td>\n      <td>Jay and Silent Bob Strike Back</td>\n      <td>[{'id': 35, 'name': 'Comedy'}]</td>\n      <td>When Jay and Silent Bob learn that their comic...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"movies_df = df[['id', 'original_title']]\nmovies_df.rename(columns={'id':'movieId'}, inplace=True)\nmovies_df.head()","metadata":{"papermill":{"duration":0.065786,"end_time":"2022-08-08T09:13:57.491741","exception":false,"start_time":"2022-08-08T09:13:57.425955","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:40.871674Z","iopub.execute_input":"2024-01-20T13:21:40.872063Z","iopub.status.idle":"2024-01-20T13:21:40.888597Z","shell.execute_reply.started":"2024-01-20T13:21:40.872031Z","shell.execute_reply":"2024-01-20T13:21:40.887197Z"},"trusted":true},"execution_count":8,"outputs":[{"execution_count":8,"output_type":"execute_result","data":{"text/plain":"   movieId               original_title\n0      862                    Toy Story\n1     8844                      Jumanji\n2    15602             Grumpier Old Men\n3    31357            Waiting to Exhale\n4    11862  Father of the Bride Part II","text/html":"<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>movieId</th>\n      <th>original_title</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>862</td>\n      <td>Toy Story</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>8844</td>\n      <td>Jumanji</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>15602</td>\n      <td>Grumpier Old Men</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>31357</td>\n      <td>Waiting to Exhale</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>11862</td>\n      <td>Father of the Bride Part II</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"ratings_df['userId'] = ratings_df['userId'].astype(str)\n\nratings = tf.data.Dataset.from_tensor_slices(dict(ratings_df[['userId', 'original_title', 'rating']]))\nmovies = tf.data.Dataset.from_tensor_slices(dict(movies_df[['original_title']]))\n\nratings = ratings.map(lambda x: {\n    \"original_title\": x[\"original_title\"],\n    \"userId\": x[\"userId\"],\n    \"rating\": float(x[\"rating\"])\n})\n\nmovies = movies.map(lambda x: x[\"original_title\"])","metadata":{"_kg_hide-output":true,"papermill":{"duration":0.202702,"end_time":"2022-08-08T09:13:57.740317","exception":false,"start_time":"2022-08-08T09:13:57.537615","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:40.890307Z","iopub.execute_input":"2024-01-20T13:21:40.890904Z","iopub.status.idle":"2024-01-20T13:21:41.088988Z","shell.execute_reply.started":"2024-01-20T13:21:40.890857Z","shell.execute_reply":"2024-01-20T13:21:41.087765Z"},"trusted":true},"execution_count":9,"outputs":[]},{"cell_type":"code","source":"print('Total Data: {}'.format(len(ratings)))\n\ntf.random.set_seed(42)\nshuffled = ratings.shuffle(100_000, seed=42, reshuffle_each_iteration=False)\n\ntrain = ratings.take(35_000)\ntest = ratings.skip(35_000).take(8_188)","metadata":{"papermill":{"duration":0.061469,"end_time":"2022-08-08T09:13:57.847529","exception":false,"start_time":"2022-08-08T09:13:57.78606","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:41.093319Z","iopub.execute_input":"2024-01-20T13:21:41.093715Z","iopub.status.idle":"2024-01-20T13:21:41.118037Z","shell.execute_reply.started":"2024-01-20T13:21:41.093682Z","shell.execute_reply":"2024-01-20T13:21:41.116663Z"},"trusted":true},"execution_count":10,"outputs":[{"name":"stdout","text":"Total Data: 44875\n","output_type":"stream"}]},{"cell_type":"code","source":"movie_titles = movies.batch(1_000)\nuser_ids = ratings.batch(1_000).map(lambda x: x[\"userId\"])\n\nunique_movie_titles = np.unique(np.concatenate(list(movie_titles)))\nunique_user_ids = np.unique(np.concatenate(list(user_ids)))\n\nprint('Unique Movies: {}'.format(len(unique_movie_titles)))\nprint('Unique users: {}'.format(len(unique_user_ids)))","metadata":{"papermill":{"duration":1.023537,"end_time":"2022-08-08T09:13:58.917115","exception":false,"start_time":"2022-08-08T09:13:57.893578","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:41.119926Z","iopub.execute_input":"2024-01-20T13:21:41.120589Z","iopub.status.idle":"2024-01-20T13:21:42.162517Z","shell.execute_reply.started":"2024-01-20T13:21:41.120541Z","shell.execute_reply":"2024-01-20T13:21:42.161507Z"},"trusted":true},"execution_count":11,"outputs":[{"name":"stdout","text":"Unique Movies: 42373\nUnique users: 671\n","output_type":"stream"}]},{"cell_type":"code","source":"class MovieModel(tfrs.models.Model):\n\n    def __init__(self, rating_weight: float, retrieval_weight: float) -> None:\n        # We take the loss weights in the constructor: this allows us to instantiate\n        # several model objects with different loss weights.\n\n        super().__init__()\n\n        embedding_dimension = 64\n\n        # User and movie models.\n        self.movie_model: tf.keras.layers.Layer = tf.keras.Sequential([\n          tf.keras.layers.StringLookup(\n            vocabulary=unique_movie_titles, mask_token=None),\n          tf.keras.layers.Embedding(len(unique_movie_titles) + 1, embedding_dimension)\n        ])\n        self.user_model: tf.keras.layers.Layer = tf.keras.Sequential([\n          tf.keras.layers.StringLookup(\n            vocabulary=unique_user_ids, mask_token=None),\n          tf.keras.layers.Embedding(len(unique_user_ids) + 1, embedding_dimension)\n        ])\n\n        # A small model to take in user and movie embeddings and predict ratings.\n        # We can make this as complicated as we want as long as we output a scalar\n        # as our prediction.\n        self.rating_model = tf.keras.Sequential([\n            tf.keras.layers.Dense(256, activation=\"relu\"),\n            tf.keras.layers.Dense(128, activation=\"relu\"),\n            tf.keras.layers.Dense(64, activation=\"relu\"),\n            tf.keras.layers.Dense(1),\n        ])\n\n        # The tasks.\n        self.rating_task: tf.keras.layers.Layer = tfrs.tasks.Ranking(\n            loss=tf.keras.losses.MeanSquaredError(),\n            metrics=[tf.keras.metrics.RootMeanSquaredError()],\n        )\n        self.retrieval_task: tf.keras.layers.Layer = tfrs.tasks.Retrieval(\n            metrics=tfrs.metrics.FactorizedTopK(\n                candidates=movies.batch(128).map(self.movie_model)\n            )\n        )\n\n        # The loss weights.\n        self.rating_weight = rating_weight\n        self.retrieval_weight = retrieval_weight\n\n    def call(self, features: Dict[Text, tf.Tensor]) -> tf.Tensor:\n        # We pick out the user features and pass them into the user model.\n        user_embeddings = self.user_model(features[\"userId\"])\n        # And pick out the movie features and pass them into the movie model.\n        movie_embeddings = self.movie_model(features[\"original_title\"])\n\n        return (\n            user_embeddings,\n            movie_embeddings,\n            # We apply the multi-layered rating model to a concatentation of\n            # user and movie embeddings.\n            self.rating_model(\n                tf.concat([user_embeddings, movie_embeddings], axis=1)\n            ),\n        )\n\n\n    def compute_loss(self, features: Dict[Text, tf.Tensor], training=False) -> tf.Tensor:\n        ratings = features.pop(\"rating\")\n\n        user_embeddings, movie_embeddings, rating_predictions = self(features)\n\n        # We compute the loss for each task.\n        rating_loss = self.rating_task(\n            labels=ratings,\n            predictions=rating_predictions,\n        )\n        retrieval_loss = self.retrieval_task(user_embeddings, movie_embeddings)\n\n        # And combine them using the loss weights.\n        return (self.rating_weight * rating_loss + self.retrieval_weight * retrieval_loss)","metadata":{"papermill":{"duration":0.066511,"end_time":"2022-08-08T09:13:59.028589","exception":false,"start_time":"2022-08-08T09:13:58.962078","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:42.16408Z","iopub.execute_input":"2024-01-20T13:21:42.164722Z","iopub.status.idle":"2024-01-20T13:21:42.18817Z","shell.execute_reply.started":"2024-01-20T13:21:42.164685Z","shell.execute_reply":"2024-01-20T13:21:42.186424Z"},"trusted":true},"execution_count":12,"outputs":[]},{"cell_type":"code","source":"model = MovieModel(rating_weight=1.0, retrieval_weight=1.0)\nmodel.compile(optimizer=tf.keras.optimizers.Adagrad(0.1))\n\ncached_train = train.shuffle(100_000).batch(1_000).cache()\ncached_test = test.batch(1_000).cache()\n\nmodel.fit(cached_train, epochs=30)","metadata":{"papermill":{"duration":240.506827,"end_time":"2022-08-08T09:17:59.581321","exception":false,"start_time":"2022-08-08T09:13:59.074494","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T13:21:42.190868Z","iopub.execute_input":"2024-01-20T13:21:42.191438Z","iopub.status.idle":"2024-01-20T14:25:28.541347Z","shell.execute_reply.started":"2024-01-20T13:21:42.191387Z","shell.execute_reply":"2024-01-20T14:25:28.539433Z"},"trusted":true},"execution_count":13,"outputs":[{"name":"stdout","text":"Epoch 1/30\n35/35 [==============================] - 100s 3s/step - root_mean_squared_error: 1.5820 - factorized_top_k/top_1_categorical_accuracy: 3.4286e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0073 - factorized_top_k/top_10_categorical_accuracy: 0.0162 - factorized_top_k/top_50_categorical_accuracy: 0.0892 - factorized_top_k/top_100_categorical_accuracy: 0.1553 - loss: 6813.4513 - regularization_loss: 0.0000e+00 - total_loss: 6813.4513\nEpoch 2/30\n35/35 [==============================] - 97s 3s/step - root_mean_squared_error: 1.0096 - factorized_top_k/top_1_categorical_accuracy: 0.0011 - factorized_top_k/top_5_categorical_accuracy: 0.0171 - factorized_top_k/top_10_categorical_accuracy: 0.0388 - factorized_top_k/top_50_categorical_accuracy: 0.1797 - factorized_top_k/top_100_categorical_accuracy: 0.2952 - loss: 6450.8815 - regularization_loss: 0.0000e+00 - total_loss: 6450.8815\nEpoch 3/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.9767 - factorized_top_k/top_1_categorical_accuracy: 8.0000e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0220 - factorized_top_k/top_10_categorical_accuracy: 0.0504 - factorized_top_k/top_50_categorical_accuracy: 0.2209 - factorized_top_k/top_100_categorical_accuracy: 0.3520 - loss: 6189.5464 - regularization_loss: 0.0000e+00 - total_loss: 6189.5464\nEpoch 4/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.9531 - factorized_top_k/top_1_categorical_accuracy: 5.7143e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0247 - factorized_top_k/top_10_categorical_accuracy: 0.0580 - factorized_top_k/top_50_categorical_accuracy: 0.2550 - factorized_top_k/top_100_categorical_accuracy: 0.4025 - loss: 5985.5387 - regularization_loss: 0.0000e+00 - total_loss: 5985.5387\nEpoch 5/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.9343 - factorized_top_k/top_1_categorical_accuracy: 4.0000e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0272 - factorized_top_k/top_10_categorical_accuracy: 0.0639 - factorized_top_k/top_50_categorical_accuracy: 0.2878 - factorized_top_k/top_100_categorical_accuracy: 0.4387 - loss: 5818.4473 - regularization_loss: 0.0000e+00 - total_loss: 5818.4473\nEpoch 6/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.9185 - factorized_top_k/top_1_categorical_accuracy: 4.2857e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0316 - factorized_top_k/top_10_categorical_accuracy: 0.0749 - factorized_top_k/top_50_categorical_accuracy: 0.3109 - factorized_top_k/top_100_categorical_accuracy: 0.4594 - loss: 5685.9057 - regularization_loss: 0.0000e+00 - total_loss: 5685.9057\nEpoch 7/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.9039 - factorized_top_k/top_1_categorical_accuracy: 5.1429e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0372 - factorized_top_k/top_10_categorical_accuracy: 0.0832 - factorized_top_k/top_50_categorical_accuracy: 0.3279 - factorized_top_k/top_100_categorical_accuracy: 0.4792 - loss: 5584.0190 - regularization_loss: 0.0000e+00 - total_loss: 5584.0190\nEpoch 8/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.8902 - factorized_top_k/top_1_categorical_accuracy: 7.4286e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0409 - factorized_top_k/top_10_categorical_accuracy: 0.0887 - factorized_top_k/top_50_categorical_accuracy: 0.3414 - factorized_top_k/top_100_categorical_accuracy: 0.4982 - loss: 5505.8207 - regularization_loss: 0.0000e+00 - total_loss: 5505.8207\nEpoch 9/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.8790 - factorized_top_k/top_1_categorical_accuracy: 0.0011 - factorized_top_k/top_5_categorical_accuracy: 0.0427 - factorized_top_k/top_10_categorical_accuracy: 0.0919 - factorized_top_k/top_50_categorical_accuracy: 0.3537 - factorized_top_k/top_100_categorical_accuracy: 0.5160 - loss: 5444.4020 - regularization_loss: 0.0000e+00 - total_loss: 5444.4020\nEpoch 10/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.8769 - factorized_top_k/top_1_categorical_accuracy: 0.0017 - factorized_top_k/top_5_categorical_accuracy: 0.0434 - factorized_top_k/top_10_categorical_accuracy: 0.0946 - factorized_top_k/top_50_categorical_accuracy: 0.3655 - factorized_top_k/top_100_categorical_accuracy: 0.5343 - loss: 5394.7945 - regularization_loss: 0.0000e+00 - total_loss: 5394.7945\nEpoch 11/30\n35/35 [==============================] - 94s 3s/step - root_mean_squared_error: 0.8638 - factorized_top_k/top_1_categorical_accuracy: 0.0023 - factorized_top_k/top_5_categorical_accuracy: 0.0433 - factorized_top_k/top_10_categorical_accuracy: 0.0970 - factorized_top_k/top_50_categorical_accuracy: 0.3761 - factorized_top_k/top_100_categorical_accuracy: 0.5485 - loss: 5353.7288 - regularization_loss: 0.0000e+00 - total_loss: 5353.7288\nEpoch 12/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.8543 - factorized_top_k/top_1_categorical_accuracy: 0.0025 - factorized_top_k/top_5_categorical_accuracy: 0.0442 - factorized_top_k/top_10_categorical_accuracy: 0.0986 - factorized_top_k/top_50_categorical_accuracy: 0.3858 - factorized_top_k/top_100_categorical_accuracy: 0.5613 - loss: 5319.1025 - regularization_loss: 0.0000e+00 - total_loss: 5319.1025\nEpoch 13/30\n35/35 [==============================] - 94s 3s/step - root_mean_squared_error: 0.8463 - factorized_top_k/top_1_categorical_accuracy: 0.0024 - factorized_top_k/top_5_categorical_accuracy: 0.0449 - factorized_top_k/top_10_categorical_accuracy: 0.1009 - factorized_top_k/top_50_categorical_accuracy: 0.3939 - factorized_top_k/top_100_categorical_accuracy: 0.5719 - loss: 5289.4642 - regularization_loss: 0.0000e+00 - total_loss: 5289.4642\nEpoch 14/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.8403 - factorized_top_k/top_1_categorical_accuracy: 0.0024 - factorized_top_k/top_5_categorical_accuracy: 0.0451 - factorized_top_k/top_10_categorical_accuracy: 0.1021 - factorized_top_k/top_50_categorical_accuracy: 0.4000 - factorized_top_k/top_100_categorical_accuracy: 0.5812 - loss: 5263.7687 - regularization_loss: 0.0000e+00 - total_loss: 5263.7687\nEpoch 15/30\n35/35 [==============================] - 94s 3s/step - root_mean_squared_error: 0.8241 - factorized_top_k/top_1_categorical_accuracy: 0.0023 - factorized_top_k/top_5_categorical_accuracy: 0.0461 - factorized_top_k/top_10_categorical_accuracy: 0.1028 - factorized_top_k/top_50_categorical_accuracy: 0.4062 - factorized_top_k/top_100_categorical_accuracy: 0.5895 - loss: 5241.2435 - regularization_loss: 0.0000e+00 - total_loss: 5241.2435\nEpoch 16/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.8271 - factorized_top_k/top_1_categorical_accuracy: 0.0023 - factorized_top_k/top_5_categorical_accuracy: 0.0462 - factorized_top_k/top_10_categorical_accuracy: 0.1035 - factorized_top_k/top_50_categorical_accuracy: 0.4108 - factorized_top_k/top_100_categorical_accuracy: 0.5962 - loss: 5221.3610 - regularization_loss: 0.0000e+00 - total_loss: 5221.3610\nEpoch 17/30\n35/35 [==============================] - 92s 3s/step - root_mean_squared_error: 0.8094 - factorized_top_k/top_1_categorical_accuracy: 0.0023 - factorized_top_k/top_5_categorical_accuracy: 0.0467 - factorized_top_k/top_10_categorical_accuracy: 0.1044 - factorized_top_k/top_50_categorical_accuracy: 0.4153 - factorized_top_k/top_100_categorical_accuracy: 0.6018 - loss: 5203.6063 - regularization_loss: 0.0000e+00 - total_loss: 5203.6063\nEpoch 18/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.8087 - factorized_top_k/top_1_categorical_accuracy: 0.0021 - factorized_top_k/top_5_categorical_accuracy: 0.0471 - factorized_top_k/top_10_categorical_accuracy: 0.1056 - factorized_top_k/top_50_categorical_accuracy: 0.4196 - factorized_top_k/top_100_categorical_accuracy: 0.6072 - loss: 5187.6940 - regularization_loss: 0.0000e+00 - total_loss: 5187.6940\nEpoch 19/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7954 - factorized_top_k/top_1_categorical_accuracy: 0.0021 - factorized_top_k/top_5_categorical_accuracy: 0.0469 - factorized_top_k/top_10_categorical_accuracy: 0.1061 - factorized_top_k/top_50_categorical_accuracy: 0.4230 - factorized_top_k/top_100_categorical_accuracy: 0.6117 - loss: 5173.2989 - regularization_loss: 0.0000e+00 - total_loss: 5173.2989\nEpoch 20/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7898 - factorized_top_k/top_1_categorical_accuracy: 0.0021 - factorized_top_k/top_5_categorical_accuracy: 0.0468 - factorized_top_k/top_10_categorical_accuracy: 0.1065 - factorized_top_k/top_50_categorical_accuracy: 0.4251 - factorized_top_k/top_100_categorical_accuracy: 0.6161 - loss: 5160.2145 - regularization_loss: 0.0000e+00 - total_loss: 5160.2145\nEpoch 21/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7810 - factorized_top_k/top_1_categorical_accuracy: 0.0020 - factorized_top_k/top_5_categorical_accuracy: 0.0470 - factorized_top_k/top_10_categorical_accuracy: 0.1070 - factorized_top_k/top_50_categorical_accuracy: 0.4279 - factorized_top_k/top_100_categorical_accuracy: 0.6191 - loss: 5148.2592 - regularization_loss: 0.0000e+00 - total_loss: 5148.2592\nEpoch 22/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7727 - factorized_top_k/top_1_categorical_accuracy: 0.0019 - factorized_top_k/top_5_categorical_accuracy: 0.0471 - factorized_top_k/top_10_categorical_accuracy: 0.1073 - factorized_top_k/top_50_categorical_accuracy: 0.4301 - factorized_top_k/top_100_categorical_accuracy: 0.6214 - loss: 5137.2792 - regularization_loss: 0.0000e+00 - total_loss: 5137.2792\nEpoch 23/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7670 - factorized_top_k/top_1_categorical_accuracy: 0.0019 - factorized_top_k/top_5_categorical_accuracy: 0.0476 - factorized_top_k/top_10_categorical_accuracy: 0.1078 - factorized_top_k/top_50_categorical_accuracy: 0.4323 - factorized_top_k/top_100_categorical_accuracy: 0.6244 - loss: 5127.1533 - regularization_loss: 0.0000e+00 - total_loss: 5127.1533\nEpoch 24/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7569 - factorized_top_k/top_1_categorical_accuracy: 0.0019 - factorized_top_k/top_5_categorical_accuracy: 0.0476 - factorized_top_k/top_10_categorical_accuracy: 0.1078 - factorized_top_k/top_50_categorical_accuracy: 0.4344 - factorized_top_k/top_100_categorical_accuracy: 0.6272 - loss: 5117.7644 - regularization_loss: 0.0000e+00 - total_loss: 5117.7644\nEpoch 25/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7513 - factorized_top_k/top_1_categorical_accuracy: 0.0019 - factorized_top_k/top_5_categorical_accuracy: 0.0477 - factorized_top_k/top_10_categorical_accuracy: 0.1082 - factorized_top_k/top_50_categorical_accuracy: 0.4356 - factorized_top_k/top_100_categorical_accuracy: 0.6295 - loss: 5109.0383 - regularization_loss: 0.0000e+00 - total_loss: 5109.0383\nEpoch 26/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7421 - factorized_top_k/top_1_categorical_accuracy: 0.0020 - factorized_top_k/top_5_categorical_accuracy: 0.0477 - factorized_top_k/top_10_categorical_accuracy: 0.1084 - factorized_top_k/top_50_categorical_accuracy: 0.4370 - factorized_top_k/top_100_categorical_accuracy: 0.6314 - loss: 5100.9069 - regularization_loss: 0.0000e+00 - total_loss: 5100.9069\nEpoch 27/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7347 - factorized_top_k/top_1_categorical_accuracy: 0.0019 - factorized_top_k/top_5_categorical_accuracy: 0.0479 - factorized_top_k/top_10_categorical_accuracy: 0.1087 - factorized_top_k/top_50_categorical_accuracy: 0.4384 - factorized_top_k/top_100_categorical_accuracy: 0.6331 - loss: 5093.2877 - regularization_loss: 0.0000e+00 - total_loss: 5093.2877\nEpoch 28/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7266 - factorized_top_k/top_1_categorical_accuracy: 0.0021 - factorized_top_k/top_5_categorical_accuracy: 0.0480 - factorized_top_k/top_10_categorical_accuracy: 0.1087 - factorized_top_k/top_50_categorical_accuracy: 0.4401 - factorized_top_k/top_100_categorical_accuracy: 0.6348 - loss: 5086.1756 - regularization_loss: 0.0000e+00 - total_loss: 5086.1756\nEpoch 29/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7183 - factorized_top_k/top_1_categorical_accuracy: 0.0019 - factorized_top_k/top_5_categorical_accuracy: 0.0482 - factorized_top_k/top_10_categorical_accuracy: 0.1089 - factorized_top_k/top_50_categorical_accuracy: 0.4416 - factorized_top_k/top_100_categorical_accuracy: 0.6364 - loss: 5079.4675 - regularization_loss: 0.0000e+00 - total_loss: 5079.4675\nEpoch 30/30\n35/35 [==============================] - 93s 3s/step - root_mean_squared_error: 0.7119 - factorized_top_k/top_1_categorical_accuracy: 0.0020 - factorized_top_k/top_5_categorical_accuracy: 0.0481 - factorized_top_k/top_10_categorical_accuracy: 0.1093 - factorized_top_k/top_50_categorical_accuracy: 0.4432 - factorized_top_k/top_100_categorical_accuracy: 0.6377 - loss: 5073.1833 - regularization_loss: 0.0000e+00 - total_loss: 5073.1833\n","output_type":"stream"},{"execution_count":13,"output_type":"execute_result","data":{"text/plain":"<keras.callbacks.History at 0x7a80de554f50>"},"metadata":{}}]},{"cell_type":"code","source":"metrics = model.evaluate(cached_test, return_dict=True)\n\nprint(f\"\\nRetrieval top-100 accuracy: {metrics['factorized_top_k/top_100_categorical_accuracy']:.3f}\")\nprint(f\"Ranking RMSE: {metrics['root_mean_squared_error']:.3f}\")","metadata":{"papermill":{"duration":16.215617,"end_time":"2022-08-08T09:18:15.851948","exception":false,"start_time":"2022-08-08T09:17:59.636331","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T14:25:28.544225Z","iopub.execute_input":"2024-01-20T14:25:28.544678Z","iopub.status.idle":"2024-01-20T14:25:46.353819Z","shell.execute_reply.started":"2024-01-20T14:25:28.544642Z","shell.execute_reply":"2024-01-20T14:25:46.35246Z"},"trusted":true},"execution_count":14,"outputs":[{"name":"stdout","text":"9/9 [==============================] - 18s 2s/step - root_mean_squared_error: 1.0437 - factorized_top_k/top_1_categorical_accuracy: 7.3278e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0024 - factorized_top_k/top_10_categorical_accuracy: 0.0050 - factorized_top_k/top_50_categorical_accuracy: 0.0291 - factorized_top_k/top_100_categorical_accuracy: 0.0559 - loss: 5765.8928 - regularization_loss: 0.0000e+00 - total_loss: 5765.8928\n\nRetrieval top-100 accuracy: 0.056\nRanking RMSE: 1.044\n","output_type":"stream"}]},{"cell_type":"code","source":"def predict_movie(user, top_n=3):\n    # Create a model that takes in raw query features, and\n    index = tfrs.layers.factorized_top_k.BruteForce(model.user_model)\n    # recommends movies out of the entire movies dataset.\n    index.index_from_dataset(\n      tf.data.Dataset.zip((movies.batch(100), movies.batch(100).map(model.movie_model)))\n    )\n\n    # Get recommendations.\n    _, titles = index(tf.constant([str(user)]))\n    \n    print('Top {} recommendations for user {}:\\n'.format(top_n, user))\n    for i, title in enumerate(titles[0, :top_n].numpy()):\n        print('{}. {}'.format(i+1, title.decode(\"utf-8\")))\n\ndef predict_rating(user, movie):\n    trained_movie_embeddings, trained_user_embeddings, predicted_rating = model({\n          \"userId\": np.array([str(user)]),\n          \"original_title\": np.array([movie])\n      })\n    print(\"Predicted rating for {}: {}\".format(movie, predicted_rating.numpy()[0][0]))","metadata":{"papermill":{"duration":0.068749,"end_time":"2022-08-08T09:18:15.975448","exception":false,"start_time":"2022-08-08T09:18:15.906699","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T14:25:46.357484Z","iopub.execute_input":"2024-01-20T14:25:46.357895Z","iopub.status.idle":"2024-01-20T14:25:46.369633Z","shell.execute_reply.started":"2024-01-20T14:25:46.35786Z","shell.execute_reply":"2024-01-20T14:25:46.368462Z"},"trusted":true},"execution_count":15,"outputs":[]},{"cell_type":"markdown","source":"# Prediction","metadata":{}},{"cell_type":"markdown","source":"### UserId: 123","metadata":{}},{"cell_type":"code","source":"predict_movie(123, 5)","metadata":{"papermill":{"duration":0.367172,"end_time":"2022-08-08T09:18:16.397711","exception":false,"start_time":"2022-08-08T09:18:16.030539","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T14:25:46.371029Z","iopub.execute_input":"2024-01-20T14:25:46.372039Z","iopub.status.idle":"2024-01-20T14:25:46.720458Z","shell.execute_reply.started":"2024-01-20T14:25:46.372Z","shell.execute_reply":"2024-01-20T14:25:46.719094Z"},"trusted":true},"execution_count":16,"outputs":[{"name":"stdout","text":"Top 5 recommendations for user 123:\n\n1. Jezebel\n2. Kurz und schmerzlos\n3. The Wanderers\n4. The Greatest Story Ever Told\n5. Un long dimanche de fiançailles\n","output_type":"stream"}]},{"cell_type":"code","source":"predict_rating(123,'Minions')","metadata":{"papermill":{"duration":0.073481,"end_time":"2022-08-08T09:18:16.525934","exception":false,"start_time":"2022-08-08T09:18:16.452453","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T14:25:46.721982Z","iopub.execute_input":"2024-01-20T14:25:46.7224Z","iopub.status.idle":"2024-01-20T14:25:46.754469Z","shell.execute_reply.started":"2024-01-20T14:25:46.722361Z","shell.execute_reply":"2024-01-20T14:25:46.753131Z"},"trusted":true},"execution_count":17,"outputs":[{"name":"stdout","text":"Predicted rating for Minions: 3.5692081451416016\n","output_type":"stream"}]},{"cell_type":"markdown","source":"So, le't examine **User 123** from historical data","metadata":{"papermill":{"duration":0.05488,"end_time":"2022-08-08T09:18:16.635569","exception":false,"start_time":"2022-08-08T09:18:16.580689","status":"completed"},"tags":[]}},{"cell_type":"code","source":"ratings_df[ratings_df['userId'] == '123']","metadata":{"papermill":{"duration":0.081603,"end_time":"2022-08-08T09:18:16.771296","exception":false,"start_time":"2022-08-08T09:18:16.689693","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T14:25:46.756707Z","iopub.execute_input":"2024-01-20T14:25:46.757692Z","iopub.status.idle":"2024-01-20T14:25:46.800568Z","shell.execute_reply.started":"2024-01-20T14:25:46.757641Z","shell.execute_reply":"2024-01-20T14:25:46.798993Z"},"trusted":true},"execution_count":18,"outputs":[{"execution_count":18,"output_type":"execute_result","data":{"text/plain":"     userId  movieId  rating                date  \\\n8360    123      233     4.0 2001-07-01 20:57:06   \n8361    123      288     5.0 2001-07-01 19:32:47   \n8362    123      407     5.0 2001-07-01 20:57:57   \n8363    123      968     3.0 2001-07-01 20:59:01   \n8364    123     1968     4.0 2001-07-01 19:30:36   \n8365    123     1976     4.0 2001-07-01 19:31:51   \n8366    123     2003     4.0 2001-07-01 19:31:51   \n8367    123     2428     5.0 2001-07-01 20:57:06   \n8368    123     2502     5.0 2001-07-01 20:59:01   \n8369    123     2699     2.0 2001-07-01 20:55:42   \n8370    123     2762     5.0 2001-07-01 20:59:54   \n8371    123     2841     5.0 2001-07-01 20:59:54   \n8372    123     2959     4.0 2001-07-01 20:57:18   \n8373    123     3016     5.0 2001-07-01 20:56:40   \n8374    123     3578     3.0 2001-07-01 20:57:37   \n8375    123     4228     5.0 2001-07-01 19:31:05   \n\n                       original_title  \\\n8360                    The Wanderers   \n8361                        High Noon   \n8362              Kurz und schmerzlos   \n8363                Dog Day Afternoon   \n8364                    Fools Rush In   \n8365                          Jezebel   \n8366              Anatomie de l'enfer   \n8367     The Greatest Story Ever Told   \n8368             The Bourne Supremacy   \n8369                          Titanic   \n8370               Young and Innocent   \n8371  Un long dimanche de fiançailles   \n8372                   License to Wed   \n8373          Dr. Jekyll and Mr. Hyde   \n8374                       Der Tunnel   \n8375          La révolution française   \n\n                                                 genres  \\\n8360                      [{'id': 18, 'name': 'Drama'}]   \n8361                    [{'id': 37, 'name': 'Western'}]   \n8362  [{'id': 18, 'name': 'Drama'}, {'id': 53, 'name...   \n8363  [{'id': 80, 'name': 'Crime'}, {'id': 18, 'name...   \n8364  [{'id': 18, 'name': 'Drama'}, {'id': 35, 'name...   \n8365  [{'id': 18, 'name': 'Drama'}, {'id': 10749, 'n...   \n8366                      [{'id': 18, 'name': 'Drama'}]   \n8367  [{'id': 18, 'name': 'Drama'}, {'id': 36, 'name...   \n8368  [{'id': 28, 'name': 'Action'}, {'id': 18, 'nam...   \n8369  [{'id': 28, 'name': 'Action'}, {'id': 18, 'nam...   \n8370  [{'id': 18, 'name': 'Drama'}, {'id': 80, 'name...   \n8371                      [{'id': 18, 'name': 'Drama'}]   \n8372                     [{'id': 35, 'name': 'Comedy'}]   \n8373  [{'id': 18, 'name': 'Drama'}, {'id': 27, 'name...   \n8374           [{'id': 878, 'name': 'Science Fiction'}]   \n8375  [{'id': 18, 'name': 'Drama'}, {'id': 10752, 'n...   \n\n                                               overview  \n8360  The streets of the Bronx are owned by 60’s you...  \n8361  High Noon is about a recently freed leader of ...  \n8362  Three friends get caught in a life of major cr...  \n8363  A man robs a bank to pay for his lover's opera...  \n8364  Alex Whitman (Matthew Perry) is a designer fro...  \n8365  In 1850s Louisiana, the willfulness of a tempe...  \n8366  A man rescues a woman from a suicide attempt i...  \n8367          All-star epic retelling of Christ's life.  \n8368  When a CIA operation to purchase classified Ru...  \n8369  A story of the romances of two couples upon th...  \n8370  Derrick De Marney finds himself in a 39 Steps ...  \n8371  In 1919, Mathilde was 19 years old. Two years ...  \n8372  Newly engaged, Ben and Sadie can't wait to sta...  \n8373  Dr. Jekyll and Mr. Hyde is a 1920 horror silen...  \n8374  The engineer MacAllan designs a tunnel, which ...  \n8375  A history of the French Revolution from the de...  ","text/html":"<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>userId</th>\n      <th>movieId</th>\n      <th>rating</th>\n      <th>date</th>\n      <th>original_title</th>\n      <th>genres</th>\n      <th>overview</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>8360</th>\n      <td>123</td>\n      <td>233</td>\n      <td>4.0</td>\n      <td>2001-07-01 20:57:06</td>\n      <td>The Wanderers</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>The streets of the Bronx are owned by 60’s you...</td>\n    </tr>\n    <tr>\n      <th>8361</th>\n      <td>123</td>\n      <td>288</td>\n      <td>5.0</td>\n      <td>2001-07-01 19:32:47</td>\n      <td>High Noon</td>\n      <td>[{'id': 37, 'name': 'Western'}]</td>\n      <td>High Noon is about a recently freed leader of ...</td>\n    </tr>\n    <tr>\n      <th>8362</th>\n      <td>123</td>\n      <td>407</td>\n      <td>5.0</td>\n      <td>2001-07-01 20:57:57</td>\n      <td>Kurz und schmerzlos</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 53, 'name...</td>\n      <td>Three friends get caught in a life of major cr...</td>\n    </tr>\n    <tr>\n      <th>8363</th>\n      <td>123</td>\n      <td>968</td>\n      <td>3.0</td>\n      <td>2001-07-01 20:59:01</td>\n      <td>Dog Day Afternoon</td>\n      <td>[{'id': 80, 'name': 'Crime'}, {'id': 18, 'name...</td>\n      <td>A man robs a bank to pay for his lover's opera...</td>\n    </tr>\n    <tr>\n      <th>8364</th>\n      <td>123</td>\n      <td>1968</td>\n      <td>4.0</td>\n      <td>2001-07-01 19:30:36</td>\n      <td>Fools Rush In</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 35, 'name...</td>\n      <td>Alex Whitman (Matthew Perry) is a designer fro...</td>\n    </tr>\n    <tr>\n      <th>8365</th>\n      <td>123</td>\n      <td>1976</td>\n      <td>4.0</td>\n      <td>2001-07-01 19:31:51</td>\n      <td>Jezebel</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 10749, 'n...</td>\n      <td>In 1850s Louisiana, the willfulness of a tempe...</td>\n    </tr>\n    <tr>\n      <th>8366</th>\n      <td>123</td>\n      <td>2003</td>\n      <td>4.0</td>\n      <td>2001-07-01 19:31:51</td>\n      <td>Anatomie de l'enfer</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>A man rescues a woman from a suicide attempt i...</td>\n    </tr>\n    <tr>\n      <th>8367</th>\n      <td>123</td>\n      <td>2428</td>\n      <td>5.0</td>\n      <td>2001-07-01 20:57:06</td>\n      <td>The Greatest Story Ever Told</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 36, 'name...</td>\n      <td>All-star epic retelling of Christ's life.</td>\n    </tr>\n    <tr>\n      <th>8368</th>\n      <td>123</td>\n      <td>2502</td>\n      <td>5.0</td>\n      <td>2001-07-01 20:59:01</td>\n      <td>The Bourne Supremacy</td>\n      <td>[{'id': 28, 'name': 'Action'}, {'id': 18, 'nam...</td>\n      <td>When a CIA operation to purchase classified Ru...</td>\n    </tr>\n    <tr>\n      <th>8369</th>\n      <td>123</td>\n      <td>2699</td>\n      <td>2.0</td>\n      <td>2001-07-01 20:55:42</td>\n      <td>Titanic</td>\n      <td>[{'id': 28, 'name': 'Action'}, {'id': 18, 'nam...</td>\n      <td>A story of the romances of two couples upon th...</td>\n    </tr>\n    <tr>\n      <th>8370</th>\n      <td>123</td>\n      <td>2762</td>\n      <td>5.0</td>\n      <td>2001-07-01 20:59:54</td>\n      <td>Young and Innocent</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 80, 'name...</td>\n      <td>Derrick De Marney finds himself in a 39 Steps ...</td>\n    </tr>\n    <tr>\n      <th>8371</th>\n      <td>123</td>\n      <td>2841</td>\n      <td>5.0</td>\n      <td>2001-07-01 20:59:54</td>\n      <td>Un long dimanche de fiançailles</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>In 1919, Mathilde was 19 years old. Two years ...</td>\n    </tr>\n    <tr>\n      <th>8372</th>\n      <td>123</td>\n      <td>2959</td>\n      <td>4.0</td>\n      <td>2001-07-01 20:57:18</td>\n      <td>License to Wed</td>\n      <td>[{'id': 35, 'name': 'Comedy'}]</td>\n      <td>Newly engaged, Ben and Sadie can't wait to sta...</td>\n    </tr>\n    <tr>\n      <th>8373</th>\n      <td>123</td>\n      <td>3016</td>\n      <td>5.0</td>\n      <td>2001-07-01 20:56:40</td>\n      <td>Dr. Jekyll and Mr. Hyde</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 27, 'name...</td>\n      <td>Dr. Jekyll and Mr. Hyde is a 1920 horror silen...</td>\n    </tr>\n    <tr>\n      <th>8374</th>\n      <td>123</td>\n      <td>3578</td>\n      <td>3.0</td>\n      <td>2001-07-01 20:57:37</td>\n      <td>Der Tunnel</td>\n      <td>[{'id': 878, 'name': 'Science Fiction'}]</td>\n      <td>The engineer MacAllan designs a tunnel, which ...</td>\n    </tr>\n    <tr>\n      <th>8375</th>\n      <td>123</td>\n      <td>4228</td>\n      <td>5.0</td>\n      <td>2001-07-01 19:31:05</td>\n      <td>La révolution française</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 10752, 'n...</td>\n      <td>A history of the French Revolution from the de...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"# Get meta data for predicted movie\nindex = tfrs.layers.factorized_top_k.BruteForce(model.user_model)\n# recommends movies out of the entire movies dataset.\nindex.index_from_dataset(\n  tf.data.Dataset.zip((movies.batch(100), movies.batch(100).map(model.movie_model)))\n)\n\n# Get recommendations.\n_, titles = index(tf.constant(['123']))\npred_movies = pd.DataFrame({'original_title': [i.decode('utf-8') for i in titles[0,:5].numpy()]})\n\npred_df = pred_movies.merge(ratings_df[['original_title', 'genres', 'overview']], on='original_title', how='left')\npred_df = pred_df[~pred_df['original_title'].duplicated()]\npred_df.reset_index(drop=True, inplace=True)\npred_df.index = np.arange(1, len(pred_df)+1)\n\npred_df","metadata":{"papermill":{"duration":0.400033,"end_time":"2022-08-08T09:18:17.227312","exception":false,"start_time":"2022-08-08T09:18:16.827279","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-01-20T14:25:46.802452Z","iopub.execute_input":"2024-01-20T14:25:46.802887Z","iopub.status.idle":"2024-01-20T14:25:47.110601Z","shell.execute_reply.started":"2024-01-20T14:25:46.802848Z","shell.execute_reply":"2024-01-20T14:25:47.10922Z"},"trusted":true},"execution_count":19,"outputs":[{"execution_count":19,"output_type":"execute_result","data":{"text/plain":"                    original_title  \\\n1                          Jezebel   \n2              Kurz und schmerzlos   \n3                    The Wanderers   \n4     The Greatest Story Ever Told   \n5  Un long dimanche de fiançailles   \n\n                                              genres  \\\n1  [{'id': 18, 'name': 'Drama'}, {'id': 10749, 'n...   \n2  [{'id': 18, 'name': 'Drama'}, {'id': 53, 'name...   \n3                      [{'id': 18, 'name': 'Drama'}]   \n4  [{'id': 18, 'name': 'Drama'}, {'id': 36, 'name...   \n5                      [{'id': 18, 'name': 'Drama'}]   \n\n                                            overview  \n1  In 1850s Louisiana, the willfulness of a tempe...  \n2  Three friends get caught in a life of major cr...  \n3  The streets of the Bronx are owned by 60’s you...  \n4          All-star epic retelling of Christ's life.  \n5  In 1919, Mathilde was 19 years old. Two years ...  ","text/html":"<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>original_title</th>\n      <th>genres</th>\n      <th>overview</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>1</th>\n      <td>Jezebel</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 10749, 'n...</td>\n      <td>In 1850s Louisiana, the willfulness of a tempe...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Kurz und schmerzlos</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 53, 'name...</td>\n      <td>Three friends get caught in a life of major cr...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>The Wanderers</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>The streets of the Bronx are owned by 60’s you...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>The Greatest Story Ever Told</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 36, 'name...</td>\n      <td>All-star epic retelling of Christ's life.</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>Un long dimanche de fiançailles</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>In 1919, Mathilde was 19 years old. Two years ...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"At a glance, we can see if **User 123** love watching Drama movies most of the time. He/She also gives a good rating for that genre. In our recommendation, We give 5 more Drama movies that we expect him/her to love the movies in a similar way with the previous watched movies.\n\nIn our dataset, we don't see any Animation movies that have been watched by **User 123**. So, it's not a surprise if the estimated rating for Minions is quite low","metadata":{"papermill":{"duration":0.054507,"end_time":"2022-08-08T09:18:17.339595","exception":false,"start_time":"2022-08-08T09:18:17.285088","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"### User: 99","metadata":{}},{"cell_type":"code","source":"predict_movie(99, 5)","metadata":{"execution":{"iopub.status.busy":"2024-01-20T14:25:47.112128Z","iopub.execute_input":"2024-01-20T14:25:47.112589Z","iopub.status.idle":"2024-01-20T14:25:47.371804Z","shell.execute_reply.started":"2024-01-20T14:25:47.112552Z","shell.execute_reply":"2024-01-20T14:25:47.370407Z"},"trusted":true},"execution_count":20,"outputs":[{"name":"stdout","text":"Top 5 recommendations for user 99:\n\n1. Bonjour Tristesse\n2. Shall We Dance?\n3. Double Dragon\n4. Dracula vs. Frankenstein\n5. The Final Cut\n","output_type":"stream"}]},{"cell_type":"code","source":"predict_rating(99,'Toy Story')","metadata":{"execution":{"iopub.status.busy":"2024-01-20T14:25:47.373426Z","iopub.execute_input":"2024-01-20T14:25:47.373821Z","iopub.status.idle":"2024-01-20T14:25:47.391606Z","shell.execute_reply.started":"2024-01-20T14:25:47.373779Z","shell.execute_reply":"2024-01-20T14:25:47.390298Z"},"trusted":true},"execution_count":21,"outputs":[{"name":"stdout","text":"Predicted rating for Toy Story: 2.6160590648651123\n","output_type":"stream"}]},{"cell_type":"markdown","source":"**Lets examine User:99 from historical data**","metadata":{}},{"cell_type":"code","source":"ratings_df[ratings_df['userId'] == '99']","metadata":{"execution":{"iopub.status.busy":"2024-01-20T14:25:47.393626Z","iopub.execute_input":"2024-01-20T14:25:47.394129Z","iopub.status.idle":"2024-01-20T14:25:47.43077Z","shell.execute_reply.started":"2024-01-20T14:25:47.39406Z","shell.execute_reply":"2024-01-20T14:25:47.429685Z"},"trusted":true},"execution_count":22,"outputs":[{"execution_count":22,"output_type":"execute_result","data":{"text/plain":"     userId  movieId  rating                date          original_title  \\\n6589     99        2     2.0 2000-02-19 22:39:18                   Ariel   \n6590     99       17     3.0 1999-09-29 06:36:39                The Dark   \n6591     99       28     3.0 1999-09-29 06:36:39          Apocalypse Now   \n6592     99      105     3.0 1999-09-29 06:43:27      Back to the Future   \n6593     99      110     3.0 2000-01-10 15:51:40  Trois couleurs : Rouge   \n...     ...      ...     ...                 ...                     ...   \n6689     99     4641     5.0 2002-03-02 02:57:04        Read It and Weep   \n6690     99     4973     5.0 2002-03-02 02:56:51           Sous le Sable   \n6691     99     4993     5.0 2003-02-09 10:35:19             5 Card Stud   \n6692     99     4995     5.0 2002-03-02 02:54:22           Boogie Nights   \n6693     99     5991     5.0 2003-02-09 01:30:04         Der letzte Mann   \n\n                                                 genres  \\\n6589  [{'id': 18, 'name': 'Drama'}, {'id': 80, 'name...   \n6590  [{'id': 27, 'name': 'Horror'}, {'id': 53, 'nam...   \n6591  [{'id': 18, 'name': 'Drama'}, {'id': 10752, 'n...   \n6592  [{'id': 12, 'name': 'Adventure'}, {'id': 35, '...   \n6593  [{'id': 18, 'name': 'Drama'}, {'id': 9648, 'na...   \n...                                                 ...   \n6689  [{'id': 35, 'name': 'Comedy'}, {'id': 18, 'nam...   \n6690  [{'id': 18, 'name': 'Drama'}, {'id': 9648, 'na...   \n6691  [{'id': 28, 'name': 'Action'}, {'id': 37, 'nam...   \n6692                      [{'id': 18, 'name': 'Drama'}]   \n6693                      [{'id': 18, 'name': 'Drama'}]   \n\n                                               overview  \n6589  Taisto Kasurinen is a Finnish coal miner whose...  \n6590  Adèle and her daughter Sarah are traveling on ...  \n6591  At the height of the Vietnam war, Captain Benj...  \n6592  Eighties teenager Marty McFly is accidentally ...  \n6593  Red This is the third film from the trilogy by...  \n...                                                 ...  \n6689  A young girl turns into an A-List celebrity ov...  \n6690  When her husband goes missing at the beach, a ...  \n6691  The players in an ongoing poker game are being...  \n6692  Set in 1977, back when sex was safe, pleasure ...  \n6693  An aging doorman, after being fired from his p...  \n\n[105 rows x 7 columns]","text/html":"<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>userId</th>\n      <th>movieId</th>\n      <th>rating</th>\n      <th>date</th>\n      <th>original_title</th>\n      <th>genres</th>\n      <th>overview</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>6589</th>\n      <td>99</td>\n      <td>2</td>\n      <td>2.0</td>\n      <td>2000-02-19 22:39:18</td>\n      <td>Ariel</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 80, 'name...</td>\n      <td>Taisto Kasurinen is a Finnish coal miner whose...</td>\n    </tr>\n    <tr>\n      <th>6590</th>\n      <td>99</td>\n      <td>17</td>\n      <td>3.0</td>\n      <td>1999-09-29 06:36:39</td>\n      <td>The Dark</td>\n      <td>[{'id': 27, 'name': 'Horror'}, {'id': 53, 'nam...</td>\n      <td>Adèle and her daughter Sarah are traveling on ...</td>\n    </tr>\n    <tr>\n      <th>6591</th>\n      <td>99</td>\n      <td>28</td>\n      <td>3.0</td>\n      <td>1999-09-29 06:36:39</td>\n      <td>Apocalypse Now</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 10752, 'n...</td>\n      <td>At the height of the Vietnam war, Captain Benj...</td>\n    </tr>\n    <tr>\n      <th>6592</th>\n      <td>99</td>\n      <td>105</td>\n      <td>3.0</td>\n      <td>1999-09-29 06:43:27</td>\n      <td>Back to the Future</td>\n      <td>[{'id': 12, 'name': 'Adventure'}, {'id': 35, '...</td>\n      <td>Eighties teenager Marty McFly is accidentally ...</td>\n    </tr>\n    <tr>\n      <th>6593</th>\n      <td>99</td>\n      <td>110</td>\n      <td>3.0</td>\n      <td>2000-01-10 15:51:40</td>\n      <td>Trois couleurs : Rouge</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 9648, 'na...</td>\n      <td>Red This is the third film from the trilogy by...</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      <td>...</td>\n    </tr>\n    <tr>\n      <th>6689</th>\n      <td>99</td>\n      <td>4641</td>\n      <td>5.0</td>\n      <td>2002-03-02 02:57:04</td>\n      <td>Read It and Weep</td>\n      <td>[{'id': 35, 'name': 'Comedy'}, {'id': 18, 'nam...</td>\n      <td>A young girl turns into an A-List celebrity ov...</td>\n    </tr>\n    <tr>\n      <th>6690</th>\n      <td>99</td>\n      <td>4973</td>\n      <td>5.0</td>\n      <td>2002-03-02 02:56:51</td>\n      <td>Sous le Sable</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 9648, 'na...</td>\n      <td>When her husband goes missing at the beach, a ...</td>\n    </tr>\n    <tr>\n      <th>6691</th>\n      <td>99</td>\n      <td>4993</td>\n      <td>5.0</td>\n      <td>2003-02-09 10:35:19</td>\n      <td>5 Card Stud</td>\n      <td>[{'id': 28, 'name': 'Action'}, {'id': 37, 'nam...</td>\n      <td>The players in an ongoing poker game are being...</td>\n    </tr>\n    <tr>\n      <th>6692</th>\n      <td>99</td>\n      <td>4995</td>\n      <td>5.0</td>\n      <td>2002-03-02 02:54:22</td>\n      <td>Boogie Nights</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>Set in 1977, back when sex was safe, pleasure ...</td>\n    </tr>\n    <tr>\n      <th>6693</th>\n      <td>99</td>\n      <td>5991</td>\n      <td>5.0</td>\n      <td>2003-02-09 01:30:04</td>\n      <td>Der letzte Mann</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>An aging doorman, after being fired from his p...</td>\n    </tr>\n  </tbody>\n</table>\n<p>105 rows × 7 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"# Get meta data for predicted movie\nindex = tfrs.layers.factorized_top_k.BruteForce(model.user_model)\n# recommends movies out of the entire movies dataset.\nindex.index_from_dataset(\n  tf.data.Dataset.zip((movies.batch(100), movies.batch(100).map(model.movie_model)))\n)\n\n# Get recommendations.\n_, titles = index(tf.constant(['99']))\npred_movies = pd.DataFrame({'original_title': [i.decode('utf-8') for i in titles[0,:5].numpy()]})\n\npred_df = pred_movies.merge(ratings_df[['original_title', 'genres', 'overview']], on='original_title', how='left')\npred_df = pred_df[~pred_df['original_title'].duplicated()]\npred_df.reset_index(drop=True, inplace=True)\npred_df.index = np.arange(1, len(pred_df)+1)\n\npred_df","metadata":{"execution":{"iopub.status.busy":"2024-01-20T14:25:47.432727Z","iopub.execute_input":"2024-01-20T14:25:47.433651Z","iopub.status.idle":"2024-01-20T14:25:47.860584Z","shell.execute_reply.started":"2024-01-20T14:25:47.433601Z","shell.execute_reply":"2024-01-20T14:25:47.859554Z"},"trusted":true},"execution_count":23,"outputs":[{"execution_count":23,"output_type":"execute_result","data":{"text/plain":"             original_title  \\\n1         Bonjour Tristesse   \n2           Shall We Dance?   \n3             Double Dragon   \n4  Dracula vs. Frankenstein   \n5             The Final Cut   \n\n                                              genres  \\\n1                      [{'id': 18, 'name': 'Drama'}]   \n2  [{'id': 18, 'name': 'Drama'}, {'id': 10749, 'n...   \n3  [{'id': 28, 'name': 'Action'}, {'id': 12, 'nam...   \n4  [{'id': 27, 'name': 'Horror'}, {'id': 878, 'na...   \n5  [{'id': 28, 'name': 'Action'}, {'id': 18, 'nam...   \n\n                                            overview  \n1  Cecile, decadent young girl who lives with her...  \n2  Upon first sight of a beautiful instructor, a ...  \n3  Set 15 years in the future in post-earthquake ...  \n4  Dracula conspires with a mad doctor to resurre...  \n5  Mysterious bomber is planting explosive device...  ","text/html":"<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>original_title</th>\n      <th>genres</th>\n      <th>overview</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>1</th>\n      <td>Bonjour Tristesse</td>\n      <td>[{'id': 18, 'name': 'Drama'}]</td>\n      <td>Cecile, decadent young girl who lives with her...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Shall We Dance?</td>\n      <td>[{'id': 18, 'name': 'Drama'}, {'id': 10749, 'n...</td>\n      <td>Upon first sight of a beautiful instructor, a ...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Double Dragon</td>\n      <td>[{'id': 28, 'name': 'Action'}, {'id': 12, 'nam...</td>\n      <td>Set 15 years in the future in post-earthquake ...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Dracula vs. Frankenstein</td>\n      <td>[{'id': 27, 'name': 'Horror'}, {'id': 878, 'na...</td>\n      <td>Dracula conspires with a mad doctor to resurre...</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>The Final Cut</td>\n      <td>[{'id': 28, 'name': 'Action'}, {'id': 18, 'nam...</td>\n      <td>Mysterious bomber is planting explosive device...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"At a glance, we can see if **User 99** love watching Drama, Action genre movies most of the time. He/She also gives a good rating for that genre. In our recommendation, We give 5 more Drama movies that we expect him/her to love the movies in a similar way with the previous watched movies.\n\nIn our dataset, we don't see any Animation movies that have been watched by **User 99**. So, it's not a surprise if the estimated rating for Toy Story is quite low","metadata":{}},{"cell_type":"markdown","source":"<h1 style=\"font-family: Trebuchet MS; font-size: 60px; color: #1b4332; text-align: center;\"><b>THE END</b></h1>","metadata":{"papermill":{"duration":0.054898,"end_time":"2022-08-08T09:18:17.559199","exception":false,"start_time":"2022-08-08T09:18:17.504301","status":"completed"},"tags":[]}}]}