GNN方法和模型的Pytorch实现。Pytorch implementation of GNN.

A PyG-based package of spectral GNNs with benchmark evaluations (SIGMOD 2026).

This project detects Pneumonia from Chest X-ray images using deep learning. It explores four approaches: A custom CNN built from scratch ResNet with transfer learning Graph Neural Network (GAT) for spatial reasoning A Multimodal model combining global and patch features The dataset used is from Kaggle's Chest X-Ray Pneumonia collection.

Graph-Based Reinforcement Learning for Structural Cross-Sectional Design Optimization

Spatio-Temporal GNN (T-GCN, A3T-GCN) for stock trend classification and price forecasting for portfolio selection.

Implementation in Pytorch of the article "E(n) Equivariant Graph Neural Network" - Satorras et al.

Master's Thesis developed at Technische Universität München (TUM) under supervision of Prof. Stephan Günnemann.

蔡老師的課程作業

Introduction to applied machine learning

Research project exploring AI agents in the Tetris environment using Reinforcement Learning and Deep Neural Networks.