3 Repos
Methods for processing multiple graph structures in a single computational batch to improve throughput.
Distinct from Batch Processing: Focuses on batching non-uniform graph structures for neural networks, distinct from matrix or query batching.
Explore 3 awesome GitHub repositories matching artificial intelligence & ml · Graph Batching Optimizations. Refine with filters or upvote what's useful.
PyTorch Geometric is a library for building and training machine learning models on graph-structured data. It provides a framework for developing graph neural networks, including a specialized system for implementing node-to-node information exchange via customizable message passing, aggregation, and update functions. The library includes a sparse data processing toolkit that utilizes accelerated CPU and CUDA kernels to perform efficient reductions on large sparse datasets. It supports the creation of specialized architectures for structured data such as 3D meshes and point clouds. The proje
Implements graph batching optimizations to group multiple small graphs into a single disjoint graph for GPU processing.
DGL is a Python library for building and training graph neural networks. It functions as a graph message passing framework and a geometric deep learning tool, enabling the development of models that analyze graph-structured data. The library is designed for large-scale graph processing, utilizing distributed training and neighbor sampling to handle datasets with billions of edges. It provides specialized support for heterogeneous graph modeling, allowing for the representation of complex real-world entities with multiple node and edge types. Its capabilities cover a wide range of graph tasks
Processes multiple small graphs in a single batch to increase computational efficiency during classification.
This is a graph convolutional network library designed for performing node and graph classification on graph-structured data. It functions as a framework for generating graph embeddings and implementing spectral convolutional neural networks to predict labels for nodes and entire graph structures. The library provides specialized tools for spectral graph convolutions, utilizing Chebyshev polynomial approximations to perform feature aggregation. It includes a multi-graph processing framework that manages batches of different graph instances through block-diagonal adjacency matrices and pooling
Provides optimized block-diagonal adjacency matrices to batch multiple graphs of different sizes for GPU processing.