Course Overview
This workshop teaches you techniques for data-parallel deep learning training on multiple GPUs to shorten the training time required for data-intensive applications. Working with deep learning tools, frameworks, and workflows to perform neural network training, you’ll learn how to decrease model training time by distributing data to multiple GPUs, while retaining the accuracy of training on a single GPU.
Course Objectives
- Understand how data parallel deep learning training is performed using multiple GPUs
- Achieve maximum throughput when training, for the best use of multiple GPUs
- Distribute training to multiple GPUs using Pytorch
- Distributed Data Parallel
- Understand and utilize algorithmic considerations specific to multi-GPU training performance and accuracy
Who Should Attend?
Experienced Python Developers
- Top-rated instructors: Our crew of subject matter experts have an average instructor rating of 4.8 out of 5 across thousands of reviews.
- Authorized content: We maintain more than 35 Authorized Training Partnerships with the top players in tech, ensuring your course materials contain the most relevant and up-to date information.
- Interactive classroom participation: Our virtual training includes live lectures, demonstrations and virtual labs that allow you to participate in discussions with your instructor and fellow classmates to get real-time feedback.
- Post Class Resources: Review your class content, catch up on any material you may have missed or perfect your new skills with access to resources after your course is complete.
- Private Group Training: Let our world-class instructors deliver exclusive training courses just for your employees. Our private group training is designed to promote your team’s shared growth and skill development.
- Tailored Training Solutions: Our subject matter experts can customize the class to specifically address the unique goals of your team.
Agenda
- Introduction
- Stochastic Gradient Descent and the Effects of Batch Size
- Training on Multiple GPUs with PyTorch Distributed Data Parallel (DDP)
- Maintaining Model Accuracy when Scaling to Multiple GPUs
- Workshop Assessment
- Final Review