rec_pangu is a flexible open-source project for recommendation systems. It incorporates diverse AI models like ranking algorithms, sequence recall, multi-interest models, and graph-based techniques. Designed for both beginners and advanced users, it enables rapid construction of efficient, custom recommendation engines.

The source code of MacGNN, The Web Conference 2024.

[CIKM 2024] Do We Really Need Graph Convolution During Training? Light Post-Training Graph-ODE for Efficient Recommendation

This project uses graph convolutional neural networks to generate embeddings, and then uses Milvus retrieval to implement a recommendation system. It provides flask services and a front-end interface.

The official implementation of Symmetric Graph Contrastive Learning against Noisy Views for Recommendation

The source code of NRCGI (Non-Recursive Cluster-Scale Graph Interacted Model for Click-Through Rate Prediction, CIKM2023).

This is the official Pytorch-version code of FlatGCN (Flattened Graph Convolutional Networks for Recommendation, DLP-KDD 2022).