ModelOps: In the driving seat of the model lifecycle 

These are great times for being a data scientist. More and more companies of all sizes and industries are turning towards model-driven business models as this provides competitive advantages. But often organizational structures are not keeping pace. Friction arises where production IT meets the analytical space and deploying an analytical model becomes a challenge, causing severe delays and high costs. In the area of software development, these issues are well known and DevOps has proven to be a good response.
Enter ModelOps, the adaption of DevOps to analytics. This session demonstrates how to realize the key idea of ModelOps: to put the developer into the driver's seat by giving more control on the model lifecycle. ModelOps is all about automation and toolchains. Git, Jenkins, Docker, Kubernetes typically are the building blocks of these toolchains. We will demonstrate how to fully orchestrate the model lifecycle in an open analytical ecosystem. Using a Python scoring model we will cover registering, deploying, and monitoring models in Docker containers on a Kubernetes cluster.
Download the presentation slides: www.dropbox.co...
Sponsored by SAS
+++ Code of Conduct ++++++++++++
github.com/mic...
+++ AI Suisse RU-vid Channel ++++
Sign up for our RU-vid Channel: / aisuisse
+++ deeplearning.ai ++++++++++++++
This is a Pie & AI event in partnership with DeepLearning.AI. Pie & AI is a series of DeepLearning.AI meetups independently hosted by community groups. This event is hosted by AI Suisse. In addition to signing up on Meetup.com, sign up here (forms.gle/vUVP...) if you'd like to receive a course promo code and future event updates from DeepLearning.AI.
deeplearning.ai