A fast, parallelized, memory efficient, and cache-optimized Python implementation of node2vec unsupervised graph representation learning algorithm.
Keywords: graph representation learning, unsupervised learning, graph embedding, node2vec, graph machine learning, Python
A pedestrian end-to-end detector for crowded scenes.
Keywords: pedestrian detection, non-maximal supression, computer vision, Python, PyTorch
The reference implementation for the PF-GNN algorithm for increasing the expressive power of Graph Neural Networks beyond the 1-WL test.
Keywords: graph neural networks, graph representation learning, particle filter, WL test, graph isomorphism, Python, PyTorch, PyTorch Geometric
Physics Informed Neural Networks (PINNs)
Physics Informed Neural Networks (PINNs) are neural networks that are trained to solve supervised learning tasks while respecting any given law of physics described by …
Keywords: physics, deep learning, neural networks, nonlinear partial differential equations, PDEs, Python, TensorFlow
The official reference implementation of the PPRGo graph neural network algorithm for scalable distributed inference on large graphs.
Keywords: Scalable graph neural network, graph representation learning, machine learning, node classification, Personalised PageRank, PyTorch
ProteinMPNN implements a deep learning based protein sequence design method that is widely applicable to current design challenges and shows outstanding performance in both in …
Keywords: protein sequence design, deep learning, message passing, machine learning, proteomics, Python, PyTorch
protEncoder is a Python package that encodes protein fasta files using different methods into smaller batches. In addition, it encodes Gene Ontology Annotation (GOA) using …
Keywords: proteins, biology, fasta, gene ontology annotation, GOA, Python
ProtTrans is providing state of the art pretrained language models for proteins. ProtTrans was trained on thousands of GPUs from Summit and hundreds of Google …
Keywords: protein, language models, language of life, representation learning, transformer, neural network, Python, PyTorch
A Python library implementing a handful of state-of-the-art causal discovery algorithms.
Keywords: causal machine learning, causal discovery, Python
PyGCL is an open source Python library for graph contrastive learning. It builds on PyTorch and supports both PyTorch Geometric and the Deep Graph Library. …
Keywords: graph representation learning, graph contrastive learning, unsupervised learning, graph neural networks, Python, PyTorch