Learn How To Build Machine Learning Models Faster at ODSC Europe

Attend the Virtual Europe Open Data Science Conference to learn how to apply GitOps and multi-tenancy to your machine learning and gain insight with live metrics tracking using Tensorboard and Kubeflow

Session: GitOps and Multi-Tenancy Combined for an Enterprise Data Science Experience on Kubeflow

When

British Summer Time (BST / GMT+1)

September 19th – 11:30 – 13:00

US Eastern Daylight Time (EDT / GMT-4)

September 19th – 06:30 – 08:00

US Pacific Daylight Time (PDT / GMT-7)

September 19th – 03:30 – 5:00

How To Attend

Get your tickets here

Session Details

Kubeflow is a machine learning (ML) platform built on top of Kubernetes. The Kubeflow project is dedicated to making deployments of ML workflows on Kubernetes simple, portable and scalable.

GitOps is the methodology of defining all infrastructure as declarative code and tracking it using git. In Kubeflow and Kubernetes, GitOps is a best practice to achieve immutable, reproducible infrastructure that can scale according to an organization’s needs.

In this session, you will: 1) learn how to apply GitOps in order to deploy and manage a Kubeflow cluster; 2) learn how to enable multiple users to work together on the same cluster in a secure and isolated way, with authentication and authorization best practices; 3) follow a data scientist’s journey to running a hyperparameter tuning optimization workflow; 4) scale up your workloads in a UI driven environment.

Session Outline
Lesson 1: GitOps and Declarative Infrastructure

Revisit the declarative nature of Kubernetes and apply GitOps best practices to get immutable, trackable and reproducible infrastructure. Deploy and manage Kubeflow using the GitOps methodology.

Lesson 2: Multi-User Kubeflow

Learn how Kubeflow and Kubernetes enforce authentication and authorization. Then see this knowledge applied in practice in order to enable multiple users to share the same Kubeflow cluster in a secure and isolated manner.

Lesson 3: Secure and Isolated User Workflows

Follow the steps of a data scientist deploying their pipelines in a secure and isolated manner. Learn how secrets are securely distributed and injected into the user’s environment. Try out an end-to-end user workflow right out of your Jupyter Notebook, by leveraging Kale, the easiest way to go from Notebook to Pipeline.

Background Knowledge
Attendees should be familiar with Kubernetes.

Speakers

Yannis Zarkadas

Yannis Zarkadas

Software Engineer

Yannis is a software engineer at Arrikto, working with Kubeflow and the Kubernetes sig-storage group. He loves contributing to open source projects and has authored the Cassandra Operator in Rook and the official Scylla Operator, which he is currently maintaining.

Stefano Fioravanzo

Stefano Fioravanzo

Software Engineer

Stefano Fioravanzo is a Software Engineer at Arrikto, his background is in Data Science and ML Research. He understands the value of building robust Machine Learning infrastructure and providing Data Scientist with the necessary tools to scale up their workflows. He works as a full-time contributor to Kubeflow and he is the creator of the Kubeflow Kale project which enables Jupyter Notebooks deployments to Kubeflow Pipelines.

Model Training with GPUs and Live Metrics Tracking with Tensorboard on Kubeflow

When

British Summer Time (BST / GMT+1)

September 19th – 09:45 – 11:15

US Eastern Daylight Time (EDT / GMT-4)

September 19th – 04:45 – 06:15

US Pacific Daylight Time (PDT / GMT-7)

September 19th – 01:45 – 03:15

How To Attend

Get your tickets here

Session Details

Kubeflow is the de facto standard for running Machine Learning (ML) workflows on Kubernetes. Its goal is to simplify the day-to-day operations of the data scientists and accelerate the production deployment of models.

Kubeflow comes with all of the tools and technologies that end users are accustomed to like Jupyter Notebooks, Tensorflow, and Tensorboard. It also provides intuitive UIs for managing and consuming the data of the cluster.

In this session you will: 1) learn the basics of Kubeflow, including configuring a Jupyter Notebook on a K8s cluster, 2) upload data from your local machine directly to the cluster using Kubeflow’s UIs, 3) tackle a real world ML problem using Keras and GPUs to train a dog breed identifier, 4) track and visualize training metrics using Tensorboard.

Session Outline
Lesson 1: Learn the basics of Kubeflow
Discover the different tools and services of Kubeflow. Configure a Jupyter Notebook, including injecting Object Store credentials and assigning GPUs. And all of this within Kubeflow’s UIs, without the need to access any terminal.

Lesson 2: Learn how to upload your local data to the Kubeflow cluster
You will learn how to use Kubeflow’s intuitive UIs and applications to upload files and folder from your local machine directly to the cloud with simple drag and drop mechanisms. You will also be able to navigate and play around with the data that lives inside your cluster’s volumes.

Lesson 3: Train a Keras model using GPUs
Create and train a Keras CNN model with GPUs. Given a dog image, the final model should be able to identify the dog breed reliably.

Lesson 4: Launch Tensorboard to visualize your training metrics
You will learn how to launch and use Tensorboard with Kubeflow to track and visualize your training metrics while they are generated live from the Jupyter Notebook.

Background Knowledge
Attendees should be familiar with Kubernetes, Jupyter Notebooks, and Tensorboard.

Speakers

Kimonos Sotirchos

Kimonas Sotirchos

Software Engineer

I’m a Software Engineer at Arrikto, working on storage solutions on the cloud. I love Open Source and have been a core contributor to the Kubeflow project for more than a year. I am the owner of the platform’s Jupyter infrastructure and my main goal is to improve the way users manage the lifecycle of their ML tools, like Notebooks, and data on top of Kubeflow.

I am also a mentor at the Kubeflow project at Google Summer of Code 2020 providing guidance for adding seamless support for launching Tensorboard instances.

Konstantinos Andriopoulos

Konstantinos Andriopoulos

Intern and Student at University of Athens

I am an undergraduate student at the National Technical University of Athens in the school of Electronic and Computer Engineering.
My current interests involve the constant development and optimization of emoFeatExtract: an open-source Python package and feature set for Speech Emotional Recognition.

Also, I’m currently a Google Summer of Code student for Kubeflow

Want The Presentations?