PyTorch's top priorities have always been flexibility and hackability while continuing to deliver good performance. PyTorch 2.0 introduces new compiler technologies that improve PyTorch's performance while maintaining the eager-mode development and user experience you know and love.
Shashank will introduce the new PyTorch 2.0 and provide an overview of the technology stack that powers the new torch.compile() API: TorchDynamo, AITAutograd, PrimTorch, and TorchInductor. The new compiler stack reduces training times across a wide range of workloads while being fully backward-compatible.
Talk Level: Intermediate
Code of Conduct: https://events.linuxfoundation.org/about/code-of-conduct/