https://github.com/longmaoteamtf/deep_recommenders
Deep Recommenders
https://github.com/longmaoteamtf/deep_recommenders
deep-learning multi-task-learning ranking recommendation-system retrieval
Last synced: 6 months ago
JSON representation
Deep Recommenders
- Host: GitHub
- URL: https://github.com/longmaoteamtf/deep_recommenders
- Owner: LongmaoTeamTf
- License: apache-2.0
- Created: 2020-03-24T09:59:26.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2023-07-06T22:48:21.000Z (over 2 years ago)
- Last Synced: 2025-03-29T20:09:19.767Z (7 months ago)
- Topics: deep-learning, multi-task-learning, ranking, recommendation-system, retrieval
- Language: Python
- Homepage:
- Size: 2.3 MB
- Stars: 327
- Watchers: 6
- Forks: 108
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Deep Recommenders
[](requirements.txt)
[](requirements.txt)
[](https://www.codacy.com/gh/LongmaoTeamTf/deep_recommenders/dashboard?utm_source=github.com&utm_medium=referral&utm_content=LongmaoTeamTf/deep_recommenders&utm_campaign=Badge_Grade)
[](https://github.com/LongmaoTeamTf/deep_recommenders/actions/workflows/codeql-analysis.yml)
[](https://github.com/LongmaoTeamTf/deep_recommenders/actions/workflows/continuous_integration.yml)
[](https://codecov.io/gh/LongmaoTeamTf/deep_recommenders)
[](LICENSE)
Deep Recommenders is an open-source recommendation system algorithm library
built by `tf.estimator` and `tf.keras` that the advanced APIs of TensorFlow.
🤗️ This Library mainly used for self-learning and improvement,
but also hope to help friends and classmates who are interested in the recommendation system to make progress together!
## Models
### Ranking
- **FM**
[[Estimator]](examples/train_fm_on_movielens_estimator.py)
[
*Factorization Machines, Osaka, 2010*
](https://www.csie.ntu.edu.tw/~b97053/paper/Rendle2010FM.pdf)
- **FFM**
[
*Field-aware Factorization Machines for CTR Prediction, RecSys, 2016*
](https://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf)
- **LS-PLM**
[
*Learning Piece-wise Linear Models from Large Scale Data for Ad Click Prediction, Alibaba, 2017*
](https://arxiv.org/pdf/1704.05194.pdf)
- **WDL**
[[Estimator]](examples/train_wdl_on_movielens_estimator.py)
[
*Wide & Deep Learning for Recommender Systems, Google, DLRS, 2016*
](https://arxiv.org/abs/1606.07792)
- **PNN**
[
*Product-based Neural Networks for User Response Prediction, IEEE, 2016*
](https://arxiv.org/abs/1611.00144)
- **FNN**
[[Estimator]](examples/train_fnn_on_movielens_estimator.py)
[
*Deep Learning over Multi-field Categorical Data: A Case Study on User Response Prediction, RayCloud, ECIR, 2016*
](https://arxiv.org/abs/1601.02376)
- **NFM**
[
*Neural Factorization Machines for Sparse Predictive Analytics, SIGIR, 2017*
](https://arxiv.org/pdf/1708.05027.pdf)
- **AFM**
[
*Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks, IJCAI, 2017*
](https://www.ijcai.org/proceedings/2017/0435.pdf)
- **DeepFM**
[[Estimator]](examples/train_deepfm_on_movielens_estimator.py)
[[Keras]](examples/train_deepfm_on_movielens_keras.py)
[
*DeepFM: A Factorization-Machine based Neural Network for CTR Prediction, Huawei, IJCAI, 2017*
](https://www.ijcai.org/proceedings/2017/0239.pdf)
- **DCN**
[
*Deep & Cross Network for Ad Click Predictions, Google, KDD, 2017*
](https://arxiv.org/abs/1708.05123)
- **xDeepFM**
[
*xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems, Microsoft, KDD, 2018*
](https://arxiv.org/pdf/1803.05170.pdf)
- **DIN**
[
*Deep Interest Network for Click-Through Rate Prediction, Alibaba, KDD, 2018*
](https://arxiv.org/abs/1706.06978)
- **DIEN**
[
*Deep Interest Evolution Network for Click-Through Rate Prediction, Alibaba, AAAI, 2019*
](https://arxiv.org/abs/1809.03672)
- **DLRM**
[
*Deep Learning Recommendation Model for Personalization and Recommendation Systems, Facebook, 2019*
](https://arxiv.org/abs/1906.00091)
### Retrieval
- **DSSM**
[
*Learning Deep Structured Semantic Models for Web Search using Clickthrough Data, Microsoft, CIKM, 2013*
](https://dl.acm.org/doi/10.1145/2505515.2505665)
- **YoutubeNet**
[
*Deep Neural Networks for YouTube Recommendations, Google, RecSys, 2016*
](https://static.googleusercontent.com/media/research.google.com/zh-CN//pubs/archive/45530.pdf)
- **SBCNM**
[
*Sampling-Bias-Corrected Neural Modeling for Large Corpus Item Recommendations, Google, RecSys, 2019*
](https://dl.acm.org/doi/10.1145/3298689.3346996)
- **EBR**
[
*Embedding-based Retrieval in Facebook Search, Facebook, KDD, 2020*
](https://arxiv.org/abs/2006.11632)
- **Item2Vec**
[
*Item2Vec: Neural Item Embedding for Collaborative Filtering, Microsoft, MLSP, 2016*
](https://arxiv.org/vc/arxiv/papers/1603/1603.04259v2.pdf)
- **Airbnb**
[
*Real-time Personalization using Embeddings for Search Ranking at Airbnb, Airbnb, KDD, 2018*
](https://dl.acm.org/doi/10.1145/3219819.3219885)
- **DeepWalk**
[
*DeepWalk: Online Learning of Social Representations, StonyBrook, KDD, 2014*
](https://arxiv.org/abs/1403.6652)
- **EGES**
[
*Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba, Alibaba, KDD, 2018*
](https://arxiv.org/abs/1803.02349)
- **GCN**
[[Keras]](experiments/gcn.ipynb)
[
*Semi-Supervised Classification with Graph Convolutional Networks, ICLR, 2017*
](https://arxiv.org/abs/1609.02907)
- **GraphSAGE**
[
*Inductive Representation Learning on Large Graphs, NIPS, 2017*
](https://arxiv.org/abs/1706.02216)
- **PinSage**
[
*Graph Convolutional Neural Networks for Web-Scale Recommender Systems, Pinterest, KDD, 2018*
](https://arxiv.org/abs/1806.01973)
- **IntentGC**
[
*IntentGC: a Scalable Graph Convolution Framework Fusing Heterogeneous Information for Recommendation, Alibaba, KDD, 2019*
](https://arxiv.org/abs/1907.12377)
- **GraphTR**
[
*Graph Neural Network for Tag Ranking in Tag-enhanced Video Recommendation, Tencent, CIKM, 2020*
](https://dl.acm.org/doi/abs/10.1145/3340531.3416021)
### Multi-task learning
- **MMoE**
[[Estimator]](examples/train_mmoe_on_synthetic_estimator.py)
[
*Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts, Google, KDD, 2018*
](https://dl.acm.org/doi/pdf/10.1145/3219819.3220007)
- **ESMM**
[
*Entire Space Multi-Task Model: An Effective Approach for Estimating Post-Click Conversion Rate, Alibaba, SIGIR, 2018*
](https://arxiv.org/pdf/1804.07931.pdf)
### NLP
- **Word2Vec**
[
*Distributed Representations of Words and Phrases and their Compositionality, Google, NIPS, 2013*
](https://papers.nips.cc/paper/2013/file/9aa42b31882ec039965f3c4923ce901b-Paper.pdf)
- **Transformer**
[[Keras]](experiments/transformer.ipynb)
[
*Attention Is All You Need, Google, NeurlPS, 2017*
](https://arxiv.org/abs/1706.03762)
- **BERT**
[
*BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, Google, NAACL, 2019*
](https://arxiv.org/abs/1810.04805)
## Supports
[1.15-passing]: https://img.shields.io/badge/1.15-passing-brightgreen
[1.15-failing]: https://img.shields.io/badge/1.15-failing-red
[2.0+-passing]: https://img.shields.io/badge/2.0+-passing-brightgreen
[2.3+-passing]: https://img.shields.io/badge/2.3+-passing-brightgreen
| Modules | TensorFlow |
| ------- | ---------------- |
| *deep_recommenders.estimator* | ![1.15-passing]
![2.0+-passing]
| *deep_recommenders.keras* | ![1.15-failing]
![2.3+-passing]
## License
[Apache License 2.0](LICENSE)