Welcome to TensorFlow London Meetup #26!
REGISTER HERE - https://www.eventbrite.com/e/london-meetup-deep-dive-into-tensorflow-26-tickets-76434364241
6:00 - Doors open. Networking. Drinks & Pizza
6:45 - Introduction
7:00 - 7:25 Talk 1: Introduction to Convolutional Neural Networks with Tensorflow by David Tyler
7:25 - 7:30 Q&A
7:30 - 7:55 Talk 2: Training to Explainability via GitOps with Kubeflow by Ryan Dawson
7:55 - 8.00 Q&A
Speaker: David Tyler, Managing Director at Outlier Technology
Title: Introduction to Convolutional Neural Network with Tensorflow
Bio: David is a techie at heart and has been working in software development, data analysis and data science for the last 20 years.
Abstract: Representing a significant advance in neural network design, CNNs present some interesting concepts and approaches that can be somewhat confusing at first glance. This talk aims to dig into some of those concepts and explain them in terms that reveal what’s happening behind the lines of we can so easily pull together in frameworks like Tensorflow and Keras. In particular we’ll be exploring one of the first CNN tutorials from the Tensorflow site, that builds a CNN for recognising handwriting.
Speaker: Ryan Dawson, Cloud Engineer at Seldon
Title: Training to Explainability via GitOps with Kubeflow
Bio: Ryan Dawson is a core member of the seldon open source team, providing tooling for machine learning deployments to Kubernetes (https://github.com/SeldonIO/seldon-core/). He has spent 10 years working in the Java Development scene in London across a variety of industries.
Abstract: Ryan will show models being trained and deployed for serving from kubeflow pipelines via gitops, highlighting git-powered audit trails and then walk through some of the observability features using Seldon with kubeflow, focusing on grafana-based monitoring. He will use the EFK stack for request logging and then perform explanations on the requests using the open-source library alibi. All of this behind the kubeflow gateway's istio authentication.