ABSTRACT: Facebook’s strength in AI innovation comes from its ability to quickly bring cutting-edge research into large scale production using a multi-faceted toolset. PyTorch provides flexible front end and more and more researchers start embracing PyTorch. However Python runtime is a critical problem in production environment. How to quickly ship PyTorch models to production becomes top priority. Multiple solutions have been proposed, and they all have their own pros and cons. Let’s take a deep-dive into these solutions. BIO: Lu Fang is a research scientist working with Facebook AI Platform team. He develops general purpose high performance deep learning frameworks including Caffe2 and PyTorch, which serve as the backbone of Facebook AI products, and also are widely used by researchers. He also leads the effort on project ONNX, i.e., Open Neural Network eXchange format, which bridges different AI framework together. Lu Fang has received his Ph.D. from UC Irvine, and M.S. and B.S. from Peking University. 

 Hosted by Professor Harry Xu

REFRESHMENTS at 3:45 pm, SPEAKER at 4:15 pm