Elevator Pitch
How do we use Tensorflow with serverless cloud platforms to bring the benefits of serverless (elastic scaling, low costs and no charge for idle) to the task of machine learning in the cloud?
This talk will explain how to design scalable serverless functions using pre-trained TensorFlow models.
Description
This talk will show how to use TensorFlow with serverless platforms to bring the benefits of serverless runtimes (elastic scalability, low pricing and no charge for idle) to the task of real-time machine learning in the cloud.
Attendees will learn why serverless platforms are great for machine learning in the cloud, understand the different approaches for deploying pre-trained models and learn how to architect scalable serverless functions when using TensorFlow.
Key issues covered will include loading TensorFlow libraries in serverless runtimes, tips on improving performance on cold and warm starts and how model scoring without a GPU affects throughput. Different methods for running TF models with serverless runtimes, including TensorFlow JS, will be compared and contrasted.
Developers do not need any prior experience with machine learning or serverless cloud platforms. This talk is applicable for all serverless developers interested in machine learning, rather than being restricted to a single platform or vendor.
Notes
I’m an experienced and regular conference speaker on serverless, having spoken at over twenty developer events on various serverless topics since 2015. Slides from my talks are available here: https://speakerdeck.com/jthomas/ Video recordings of my talks are available here: https://bit.ly/james_thomas
Since 2016, I’ve been blogging (http://jamesthom.as/blog/categories/serverless/), developing open-source projects (https://github.com/jthomas) and presenting at conferences about serverless cloud platforms. I’m a contributor to the Apache OpenWhisk project, an open-source serverless platform, having created (and maintaining) the official project JavaScript SDK and developed the openwhisk integration with The Serverless Framework. I also organised the London edition of the Serverless Days conference series.