Scaling Tensorflow data processing with (TF Dev Summit '20)

As model training becomes more distributed in nature, has evolved to be more distribution aware and performant. This talk presents tools for scaling TensorFlow data processing. In particular: service that allows your pipeline to run on a cluster of machines, and that materializes the results to disk for reuses across multiple invocations.

Rohan Jain – Staff Software Engineer

GitHub Distributed service → Build TensorFlow input pipelines →
Better performance with the API →
GitHub snapshot →

Watch all TensorFlow Dev Summit 2020 sessions →
Subscribe to the TensorFlow YouTube channel →

event: TensorFlow Dev Summit 2020; re_ty: Publish; product: TensorFlow – General; fullname: Rohan Jain;

Source of this TensorFlow AI Video

AI video(s) you might be interested in …