Accelerating Graph Neural Network Training on Intel CPU through Fused Sampling & Hybrid Partitioning

Intel Labs and AIA developed a new graph sampling method called “fused sampling” that achieves up to 2x speedup in training Graph Neural Networks (GNNs) on CPUs. The new sampling pipeline is now part of the Deep Graph Library (DGL), one of the most popular libraries for training GNNs.

Ce contenu a été publié dans Non classé. Vous pouvez le mettre en favoris avec ce permalien.