As models are getting larger there is a need to enable libraries and techniques that reduce the memory size to ensure the models will fit into the device's memory. Join us on Thursday, February 16th, 2023 at 9AM PST, for a live webinar, Getting Started with Habana: Deep Speed Optimization on Large Models.
In this webinar, we will review the basic steps needed to enable DeepSpeed on Gaudi and show how the ZeRO1 and ZeRO2 memory optimizers and Activation Checkpointing are used to reduce memory usage on a large model. We’ll also show you how to use Gaudi APIs to detect the peak memory of any model and provide guidance on when to use these techniques. A live Q&A to follow.
We look forward to seeing you!
