PyTorch Mobile: Past, present and 2.x future
Raziel Alvarez Guevara speaks at PyTorch Conference 2022 about PyTorch Mobile: past, present and 2.x future. PyTorch Mobile enables running PyTorch models on mobile devices, an important and growing area in the ML ecosystem that beyond individual benefits on privacy, reduced latency and resilience to connectivity issues, enables products that otherwise could not exist with server-side ML. With the growth of this space, the variety of devices and use-cases to target have increased significantly. Previous approaches to meet the often conflicting needs have focused on creating narrow solutions that significantly increase the friction of deploying the original ML model, often by leaving the original framework in which the model was authored. In this talk, we will cover our success and learnings when developing PyTorch Mobile, as well as provide an update on how we will both significantly increase the coverage of devices and use-cases we can support as well as further reduce the friction in the research to production workflow. Visit our website: https://pytorch.org/ Read our blog: https://pytorch.org/blog/ Follow us on Twitter: https://twitter.com/PyTorch Follow us on LinkedIn: https://www.linkedin.com/company/pyto... Follow us on Facebook: https://www.facebook.com/pytorch #PyTorch #ArtificialIntelligence #MachineLearning
Raziel Alvarez Guevara speaks at PyTorch Conference 2022 about PyTorch Mobile: past, present and 2.x future. PyTorch Mobile enables running PyTorch models on mobile devices, an important and growing area in the ML ecosystem that beyond individual benefits on privacy, reduced latency and resilience to connectivity issues, enables products that otherwise could not exist with server-side ML. With the growth of this space, the variety of devices and use-cases to target have increased significantly. Previous approaches to meet the often conflicting needs have focused on creating narrow solutions that significantly increase the friction of deploying the original ML model, often by leaving the original framework in which the model was authored. In this talk, we will cover our success and learnings when developing PyTorch Mobile, as well as provide an update on how we will both significantly increase the coverage of devices and use-cases we can support as well as further reduce the friction in the research to production workflow. Visit our website: https://pytorch.org/ Read our blog: https://pytorch.org/blog/ Follow us on Twitter: https://twitter.com/PyTorch Follow us on LinkedIn: https://www.linkedin.com/company/pyto... Follow us on Facebook: https://www.facebook.com/pytorch #PyTorch #ArtificialIntelligence #MachineLearning