{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.11.11","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":38760,"databundleVersionId":4493939,"sourceType":"competition"},{"sourceId":4461402,"sourceType":"datasetVersion","datasetId":2611514},{"sourceId":4474043,"sourceType":"datasetVersion","datasetId":2601572}],"dockerImageVersionId":31040,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# 🛒 A Practical Showcase of Sequential Recommendations Using GRU4Rec on the OTTO Dataset\n\nIn this notebook, cutting-edge technologies are applied to real-world data to build a **personalized recommendation system** for users on an e-commerce platform.\n\n**RecBole**, a powerful deep learning-based recommendation framework, is beign used to implement **GRU4Rec** a state-of-the-art **sequential collaborative filtering model**. This allows us to predict the next likely item a user will interact with based on their recent behavior.\n\nThe dataset powering this project is the **[OTTO Recommender Systems Dataset](https://github.com/otto-de/recsys-dataset)**, which is based on **real user interaction data** from the OTTO online retail platform. It contains millions of anonymized events such as clicks, cart additions, and purchases, and reflects realistic user behavior in an online store.\n\n---\n\n🔍 **Goals of this notebook:**\n- Preprocess and filter OTTO dataset for sequential modeling\n- Train a GRU4Rec model using RecBole and evaluate it by using ranking metrics like Recall@10 and MRR@10\n- Generate top-N product recommendations for individual user sessions","metadata":{}},{"cell_type":"markdown","source":"## 🧠 Introduction to Recommendation Systems\n\nA **recommendation system** is a machine learning model that suggests relevant content, products, or actions to users by learning from their **past behavior** and **real-time interactions**.\n\nThere are several types of recommendation systems, commonly including:\n- **Collaborative filtering**: Learns from patterns of users and items (e.g. \"users like you also liked...\")\n- **Content-based filtering**: Recommends items similar to those the user has interacted with (e.g. similar genre, brand, or features)\n- **Hybrid models**: Combine collaborative and content-based methods to overcome limitations of each approach\n\nBelow is a simplified diagram illustrating how a user interacts with different types of content (like actors, songs, or movies), and how those actions feed into the recommendation system to generate personalized suggestions:\n\n![](https://miro.medium.com/v2/resize:fit:1400/1*6gDIv9mWyc8vkdGerBKiXw.jpeg)","metadata":{}},{"cell_type":"markdown","source":"### Install RecBole for this notebook","metadata":{}},{"cell_type":"code","source":"!pip install recbole","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2025-07-07T12:50:26.853962Z","iopub.execute_input":"2025-07-07T12:50:26.854743Z","iopub.status.idle":"2025-07-07T12:51:49.389588Z","shell.execute_reply.started":"2025-07-07T12:50:26.854699Z","shell.execute_reply":"2025-07-07T12:51:49.388622Z"},"jupyter":{"outputs_hidden":true},"collapsed":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### 📂 Load and Explore the Data\n\nFor this notebook, we will be using the **smaller version** of the OTTO dataset in `.parquet` format, which is more memory-efficient and suitable for interactive analysis.\n\nTo efficiently load and work with the data, we use the **Polars** library instead of pandas. Polars offers significant performance benefits for large tabular datasets due to its Rust-based backend and support for multi-threaded operations.\n\nDue to memory restriction on Kaggle, we will be loading the last week of training data and the testing set","metadata":{}},{"cell_type":"code","source":"import polars as pl","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T15:07:11.923692Z","iopub.execute_input":"2025-07-07T15:07:11.924226Z","iopub.status.idle":"2025-07-07T15:07:11.928051Z","shell.execute_reply.started":"2025-07-07T15:07:11.924200Z","shell.execute_reply":"2025-07-07T15:07:11.927143Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# RecBole expects an atomic file, which we create here\n# more information: https://recbole.io/atomic_files.html\n\ntrain = pl.read_parquet('/kaggle/input/otto-train-and-test-data-for-local-validation/test.parquet')\ntest = pl.read_parquet('/kaggle/input/otto-full-optimized-memory-footprint/test.parquet')\n\ndf = pl.concat([train, test])\n\ndf = df.sort(['session', 'aid', 'ts'])\ndf = df.with_columns((pl.col('ts') * 1e9).alias('ts')) # unix nanoseconds are expected by RecBole\ndf = df.rename({'session': 'session:token', 'aid': 'aid:token', 'ts': 'ts:float'}) # the columnname including :[type] is expected by RecBole\n\n!mkdir /kaggle/working/recbox_data\ndf['session:token', 'aid:token', 'ts:float'].write_csv('/kaggle/working/recbox_data/recbox_data.inter', separator='\\t')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T12:51:49.925447Z","iopub.execute_input":"2025-07-07T12:51:49.925647Z","iopub.status.idle":"2025-07-07T12:51:51.910256Z","shell.execute_reply.started":"2025-07-07T12:51:49.925631Z","shell.execute_reply":"2025-07-07T12:51:51.909622Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T15:32:29.852385Z","iopub.execute_input":"2025-07-07T15:32:29.852984Z","iopub.status.idle":"2025-07-07T15:32:29.858365Z","shell.execute_reply.started":"2025-07-07T15:32:29.852961Z","shell.execute_reply":"2025-07-07T15:32:29.857641Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"For our dataset there are 3 types \"clicked\" - 0, \"cart\" - 1 and \"order\" - 2 these strings are being transformed into ints here to safe diskspace.","metadata":{}},{"cell_type":"code","source":"df.filter(pl.col(\"session:token\") == 11098530), df.filter(pl.col(\"session:token\") == 11098556), df.filter(pl.col(\"session:token\") == 12899775)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T15:32:58.768619Z","iopub.execute_input":"2025-07-07T15:32:58.769319Z","iopub.status.idle":"2025-07-07T15:32:58.777866Z","shell.execute_reply.started":"2025-07-07T15:32:58.769294Z","shell.execute_reply":"2025-07-07T15:32:58.777146Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"In these example sessions 11098530, 11098556 and 12899775: users looked at different items and acted differently. The session 11098530 resulted in a couple of item clicks but there was no order, one item has been added to the cart at the end of the session. While session 11098556 added an item to the cart the session continued by looking at different items. In contrast to the other two sessions, session 12899775 is an one click event, clicking at one item (viewing it) and ending the session afterwards.\n\nSession-based recommenders ensure that even when cookies are being declined, users can still receive relevant recommendations. These models rely solely on real-time activity within the current session such as clicks or views without requiring any long-term user history or identification. In use cases like ours, where each visit is treated as a new anonymous session, the recommender adapts instantly to session behavior, making it privacy friendly and effective for first-time or unlogged users.","metadata":{}},{"cell_type":"code","source":"import logging\nfrom logging import getLogger\nfrom recbole.config import Config\nfrom recbole.data import create_dataset, data_preparation\nfrom recbole.model.sequential_recommender import GRU4Rec\nfrom recbole.trainer import Trainer\nfrom recbole.utils import init_seed, init_logger\n\nfrom recbole.utils.case_study import full_sort_topk","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T12:51:51.961524Z","iopub.execute_input":"2025-07-07T12:51:51.961723Z","iopub.status.idle":"2025-07-07T12:52:10.677231Z","shell.execute_reply.started":"2025-07-07T12:51:51.961699Z","shell.execute_reply":"2025-07-07T12:52:10.676651Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## 🔁 GRU4Rec – Sequential Recommender with GRUs\n\nThe model we are using is **GRU4Rec**, a powerful sequential recommendation model based on **Recurrent Neural Networks (RNNs)**, specifically using **Gated Recurrent Units (GRUs)**.\n\nThe architecture begins with an **embedding layer**, which transforms each item into a dense vector representation. These embeddings are passed through **multiple GRU layers**, which capture the **sequential patterns** in a user's interactions. The model then outputs a **score for each possible item**, representing how likely it is to be the next item the user interacts with.\n\nGRU4Rec is a **collaborative filtering model**, as it learns purely from **user-item interaction sequences**. It does not use any side information (e.g. item metadata, user profiles).  \nIn real-world applications, it's often beneficial to use **hybrid models**, which combine interaction data with **content features** for improved personalization and cold-start performance.","metadata":{}},{"cell_type":"markdown","source":"![](https://www.researchgate.net/publication/324055051/figure/fig3/AS:962202051821599@1606418254544/Architecture-of-the-gru4rec-neural-network-Adapted-from-Hidasi-etal-2016a.png)","metadata":{},"attachments":{"bbd5d262-3f4a-4059-be37-4987830f1ab5.png":{"image/png":"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"},"d1ca8316-4db3-4da6-9a00-35d2f88859f2.png":{"image/png":"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"},"6b015142-8c52-458c-a120-5af6e66f2670.png":{"image/png":"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"}}},{"cell_type":"code","source":"MAX_ITEM = 20  # limit the input sequence to the last 20 items a user interacted with\n\nparameter_dict = {\n    'data_path': '/kaggle/working/',              # path where the data (e.g. inter.csv) is stored\n    'USER_ID_FIELD': 'session',                   # field used to identify users (sessions in this case)\n    'ITEM_ID_FIELD': 'aid',                       # field used to identify items\n    'TIME_FIELD': 'ts',                           # timestamp field for ordering interactions\n    'user_inter_num_interval': \"[5,Inf)\",         # keep users with at least 5 interactions\n    'item_inter_num_interval': \"[5,Inf)\",         # keep items with at least 5 interactions\n    'load_col': {'inter': ['session', 'aid', 'ts']},  # load only these columns from the interaction file\n    'train_neg_sample_args': None,                # no negative sampling (use full item ranking)\n    'epochs': 10,                                 # number of training epochs\n    'stopping_step': 3,                           # stop early if no improvement after 3 valid steps\n\n    'eval_batch_size': 1024,                      # batch size during evaluation\n    # 'train_batch_size': 1024,                   # (optional) batch size for training\n    # 'enable_amp': True,                         # (optional) enable mixed-precision training\n    'MAX_ITEM_LIST_LENGTH': MAX_ITEM,             # max number of past items used in sequence\n    'eval_args': {\n        'split': {'RS': [9, 1, 0]},               # random split: 90% train, 10% valid, 0% test\n        'group_by': 'user',                       # group data per user/session\n        'order': 'TO',                            # respect temporal order\n        'mode': 'full'                            # use full item list for evaluation\n    }\n}\n\n# initialize RecBole with the GRU4Rec model, dataset name, and custom parameters \nconfig = Config(model='GRU4Rec', dataset='recbox_data', config_dict=parameter_dict)\n\n# init random seed\ninit_seed(config['seed'], config['reproducibility'])\n\n# logger initialization\ninit_logger(config)\nlogger = getLogger()\n\n# create handlers\nc_handler = logging.StreamHandler()\nc_handler.setLevel(logging.INFO)\nlogger.addHandler(c_handler)\n\n# write config info into log\nlogger.info(config)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T12:52:10.677940Z","iopub.execute_input":"2025-07-07T12:52:10.678430Z","iopub.status.idle":"2025-07-07T12:52:11.122745Z","execution_failed":"2025-07-07T15:44:10.005Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# create the dataset based on our configurations\ndataset = create_dataset(config)\nlogger.info(dataset)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T12:57:19.922320Z","iopub.execute_input":"2025-07-07T12:57:19.923042Z","iopub.status.idle":"2025-07-07T12:58:27.519345Z","shell.execute_reply.started":"2025-07-07T12:57:19.923016Z","shell.execute_reply":"2025-07-07T12:58:27.518812Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## Training\n\nWith our data prepared and processed into sequential format, we now train the **GRU4Rec** model using the RecBole framework.\n\nGRU4Rec will learn to predict the next likely item a user will interact with, based solely on the sequence of actions within each session. This allows the model to generate personalized, session-aware recommendations without requiring long-term user tracking.","metadata":{}},{"cell_type":"code","source":"# prepare the dataset by splitting it into train, valid and test chunks\ntrain_data, valid_data, test_data = data_preparation(config, dataset)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T12:58:27.520441Z","iopub.execute_input":"2025-07-07T12:58:27.520646Z","iopub.status.idle":"2025-07-07T12:59:53.760037Z","shell.execute_reply.started":"2025-07-07T12:58:27.520629Z","shell.execute_reply":"2025-07-07T12:59:53.759512Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# model loading and initialization\nmodel = GRU4Rec(config, train_data.dataset).to(config['device'])\nlogger.info(model)\n\n# trainer loading and initialization\ntrainer = Trainer(config, model)\n\n# model training\nbest_valid_score, best_valid_result = trainer.fit(train_data, valid_data)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T12:59:53.760640Z","iopub.execute_input":"2025-07-07T12:59:53.760839Z","iopub.status.idle":"2025-07-07T13:47:05.072755Z","shell.execute_reply.started":"2025-07-07T12:59:53.760810Z","shell.execute_reply":"2025-07-07T13:47:05.072104Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Evaluation Metrics Explained\n\n- **Recall@10**: Measures the proportion of relevant items successfully retrieved in the top 10 recommendations.\n\n- **MRR@10 (Mean Reciprocal Rank)**: Evaluates how highly the first relevant item is ranked in the top 10 list. A higher score indicates the relevant item appears earlier.\n\n- **NDCG@10 (Normalized Discounted Cumulative Gain)**: Considers both the relevance and the rank position of items, giving more credit when relevant items are ranked higher.\n\n- **Hit@10**: Indicates whether at least one relevant item appears in the top 10. A simple check for presence, regardless of position.\n\n- **Precision@10**: Calculates the proportion of the top 10 recommended items that are actually relevant.\n","metadata":{}},{"cell_type":"markdown","source":"In this case, our model continues to improve steadily during training, whereas models often begin to lose momentum or overfit in later stages.\nIn a real-world scenario, it is recommended to increase the number of training epochs and rely on early stopping based on validation performance. This ensures that the best-performing model is selected automatically, without the need to guess the optimal stopping point manually. We could further improve the model's performance using techniques like a learning rate scheduler or tuning the model’s dimensionality. However, for showcase purposes, this setup is sufficient and provides a solid baseline.","metadata":{}},{"cell_type":"markdown","source":"## Inference\n\nNow that the model is trained, we will define a function that recommends the top 10 items for a given user session.\nUsing a session_id, the function will analyze the user's interaction history within that session and predict which items are most relevant to recommend next.","metadata":{}},{"cell_type":"code","source":"import torch\nfrom recbole.data.interaction import Interaction\nfrom recbole.utils.case_study import full_sort_scores\n\ndef recommend_for_sessions(external_session_id, model, dataset, top_k=10):\n     # Convert external session ID to internal\n    internal_session_id = dataset.token2id('session', external_session_id)\n\n    # load interaction data\n    inter_feat = dataset.inter_feat\n    inter_df = pl.DataFrame({\n        'session': inter_feat['session'].tolist(),\n        'aid': inter_feat['aid'].tolist(),\n        'ts': inter_feat['ts'].tolist(),\n    })\n\n    # filter and sort session history\n    session_history = (\n        inter_df\n        .filter(pl.col('session') == internal_session_id)\n        .sort('ts')\n    )\n    internal_item_ids = session_history['aid'].to_list()\n\n    # prepare interaction\n    item_list_field = dataset.iid_field + '_list'\n    user_field = dataset.uid_field\n\n    interaction = Interaction({\n        user_field: torch.tensor([internal_session_id]),\n        item_list_field: torch.tensor([internal_item_ids]),\n        'item_length': torch.tensor([len(internal_item_ids)]),\n    })\n\n    # predict scores\n    model.eval()\n    scores = model.full_sort_predict(interaction.to(model.device))\n    top_k_indices = torch.topk(scores[0], k=top_k).indices.tolist()\n    external_item_ids = dataset.id2token(dataset.iid_field, top_k_indices)\n\n    # print result\n    print(f\"Top {top_k} recommended items for session {external_session_id}:\")\n    for internal_id, external_id in zip(top_k_indices, external_item_ids):\n        print(f\"  Internal ID: {internal_id}  →  External ID: {external_id}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T14:38:46.118458Z","iopub.execute_input":"2025-07-07T14:38:46.119300Z","iopub.status.idle":"2025-07-07T14:38:46.125566Z","shell.execute_reply.started":"2025-07-07T14:38:46.119275Z","shell.execute_reply":"2025-07-07T14:38:46.124938Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Please note, that RecBole maps our external article/session ids e.g. [1855508, 535924, 214372, ...] to internal ids e.g. [0, 1, 2, ...]\n\nTo make that clear, I'm printing both ids, which still represents the same article","metadata":{}},{"cell_type":"code","source":"recommend_for_sessions('11098556', model, dataset, top_k=10)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T14:38:51.048261Z","iopub.execute_input":"2025-07-07T14:38:51.048714Z","iopub.status.idle":"2025-07-07T14:38:51.596482Z","shell.execute_reply.started":"2025-07-07T14:38:51.048691Z","shell.execute_reply":"2025-07-07T14:38:51.595945Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"recommend_for_sessions('11098530', model, dataset, top_k=10)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T14:39:29.271639Z","iopub.execute_input":"2025-07-07T14:39:29.272274Z","iopub.status.idle":"2025-07-07T14:39:29.843633Z","shell.execute_reply.started":"2025-07-07T14:39:29.272247Z","shell.execute_reply":"2025-07-07T14:39:29.843134Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"recommend_for_sessions('11098533', model, dataset, top_k=10)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-07T14:44:00.271662Z","iopub.execute_input":"2025-07-07T14:44:00.272134Z","iopub.status.idle":"2025-07-07T14:44:00.815360Z","shell.execute_reply.started":"2025-07-07T14:44:00.272108Z","shell.execute_reply":"2025-07-07T14:44:00.814846Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Based on the session history of session 11098533, the model predicts that the most relevant next items to recommend are (in order of highest to lowest predicted score):\n1074173, 978918, 1622419, ...\n\nThese are the external item IDs, representing the top-ranked items according to the model’s confidence.","metadata":{}}]}