How to Speed up Deep Learning with GPUs

OVERFIT: AI, Machine Learning, and Deep Learning Made Simple

We discuss how to utilize the processing power of Graphics Processing Units (GPUs) to speed up deep learning calculations, particularly in the context of training neural networks. It outlines how to assign data to different GPUs to minimize data transfer times, a crucial aspect of performance optimization. The text highlights the importance of understanding device allocation and data movement to maximize efficiency. It also provides practical examples and code snippets demonstrating how to store data on GPUs, copy data between GPUs, and utilize GPUs within neural networks. The text concludes by summarizing the key points and posing several exercises to challenge the reader's understanding of the discussed concepts.

Read more: https://d2l.ai/chapter_builders-guide/use-gpu.html