An Industrial Graph Neural Network Framework

Programmable CUDA/C++ GPU Graph Analytics

🍇 A C++ library for parallel graph processing (GRAPE) 🍇

This shows a basic implementation of the global nearest neighbour (GNN) multi target Tracker. Kalman filter is used for Tracking and Auction Algorithm for determining the assignment of measurments to filters.

❤️ CUDA/C++ GPU graph analytics simplified.

Dorylus: Affordable, Scalable, and Accurate GNN Training

High performance RDMA-based distributed feature collection component for training GNN model on EXTREMELY large graph

Code for Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture (accepted by PVLDB).The outdated write-up (https://arxiv.org/abs/2101.07956) explains engineering details, but only a portion of the functionality is migrated to this newer PyTorch version 1.8.0nightly (e152ca5).

[RSS 2024] AdaptiGraph: Material-Adaptive Graph-Based Neural Dynamics for Robotic Manipulation

PySCF on IPU

FPLearner is a DL-based tool to predict performance and accuracy of mixed-precision programs that can be used in dynamic precision tuners to save time.

GeoT: Tensor Centric Library for Graph Neural Network via Efficient Segment Reduction on GPU

A C++ library for the creation of a large dataset of amino acid sidechain perturbations, own PDB Parser code included and some other things related.

HET: The HET Hetero-GNN Kernel Optimization and Code Generation Project