GTP-ViT

The official GTP-ViT code for token summarisation for efficient vision transformers via graph-based token propagation.

Keywords: vision transformer, graph neural network, computer vision, machine learning, transformer architecture, deep learning, Python, PyTorch

Half-Hop

Half-Hop is a graph augmentation technique that up-samples a graph by adding nodes to slow down message passing reducing the effects of over-smoothing in graph …

Keywords: graph augmentation, over-smoothing, message passing graph neural networks, graph representation learning, GNNs, Python, PyTorch

hedwig

Hedwig is Python library implementing several deep learning models for document classification.

Keywords: document classification, natural language processing, deep learning, PyTorch, NLP, DocBERT, KimCNN, HAN, RegLSTM, XMLCNN

Hidden Fluid Mechanics (HFM)

Hidden Fluid Mechanics (HFM) is a physics informed deep learning framework capable of encoding an important class of physical laws governing fluid motions, namely the …

Keywords: physics, deep learning, fluid dynamics, Python, TensorFlow

IGCM

The reference implementation for the Inductive Graph-based Matrix Completion algorithm. IGMC is an inductive matrix completion model based on graph neural networks without using any …

Keywords: Inductive Graph-based Matrix Completion, graph neural networks, GNN, geometric deep learning

JAX

JAX is a Python library for accelerator-oriented array computation and program transformation, designed for high-performance numerical computing and large-scale machine learning.

Keywords: differentiable programs, neural networks, JIT, machine learning, deep learning, Python

Jeometric

A Python library for Graph Neural Networks (GNNs) in JAX.

Keywords: graph neural networks, GNN, machine learning, GCN, Python, JAX

jraph

jraph is a graph neural network (for message passing GNNs) library for JAX. It is inspired by the TensorFlow library graph_nets and has support for …

Keywords: graph machine learning, graph neural networks, GNN, geometric deep learning, machine learning, JAX

Karate Club

Karate Club is an unsupervised machine learning extension library for NetworkX.

Karate Club consists of state-of-the-art methods to do unsupervised learning on graph structured data. …

Keywords: Karate Club, network embedding, graph machine learning, community detection, Python, NetworkX

LGNpy

A small Python library for working with linear Gaussian Bayesian Networks that includes learning parameters and inference with evidence.

Keywords: Bayesian network, graphical models, linear Gaussian Bayesian Network, belief propagation, Python