On May 18th we hosted the “Intro to Kubeflow: Katib Training and Certification” prep course. In this blog post we’ll recap some highlights, plus give a summary of the Q&A from the class.
Congratulations to Samrat Kadge!
The first attendee to earn the “Katib” certificate at the conclusion of the course was Samrat Kadge! A free MiniKF hoodie and shirt are on the way, well done!
First, thanks for voting for your favorite charity!
With the unprecedented circumstances facing our global community, Arrikto is looking for even more ways to contribute. With this in mind, we thought that in lieu of swag we could give course attendees the opportunity to vote for their favorite charity and help guide our monthly donation to charitable causes. The charity that won this workshop’s voting was Every Mother Counts (EMC). The Every Mother Counts is a non-profit organization dedicated to making pregnancy and childbirth safe for every mother. They inform, engage, and mobilize new audiences to take actions and raise funds that support maternal health programs around the world. We are pleased to be making a donation of $100 to them on behalf of the Kubeflow community. Again, thanks to all of you who attended and voted!
What topics were covered in the course?
This initial course aimed to get data scientists and DevOps engineers with little or no experience familiar with the fundamentals of how Kubeflow works.
- Kubeflow Fundamentals Review
- Katib Basics
- Getting Started with Katib
- Navigating the Katib UI
- Running Katib Experiments
- Kale Basics
- AutoML Example: Kaggle’s OpenVaccine Notebook
- AutoML Example: Kaggle’s Bulldozers Notebook
What did I miss?
Here’s a short teaser from the 90 minute training. In this video we show you how to use Katib and Kale inside of Kubeflow to perform hyperparameter tuning on Kaggle’s OpenVaccine notebook.
Missed the May 18 Kubeflow Pipelines training?
If you were unable to join us last week, you can sign up for upcoming Fundamentals, Notebooks, Pipelines and Kale/Katib courses here.
NEW: Advanced Kubeflow, Notebooks and Pipelines Workshops
We are excited to announce a new series of FREE workshops focused on taking popular Kaggle and Udacity machine learning examples from “Notebook to Pipeline.” Registration is now open for the following workshops:
- May 25: The Kaggle Blue Book for Bulldozers Machine Learning Example
- May 26: Distributed Training Workshop
- Jun 1: Training course: Kubeflow Fundamentals
- Jun 8: The Kaggle Udacity Dog Breed Classification Example
- Jun 9: From Kubernetes to Kubeflow Workshop
- Jun 15: Training course: Jupyter Notebooks Fundamentals
If you are ready to put what you’ve learned into practice with hands-on labs? Then check out Arrikto Academy! On this site you’ll find a variety of FREE skills-building labs and tutorials including:
- Kubeflow Use Cases – Kaggle OpenVaccine, Kaggle Titanic Disaster
- Kubeflow Functionality – Kale, Katib
- Enterprise Kubeflow Skills – Kale SDK
Q&A from the training
Below is a summary of some of the questions that popped into the Q&A box during the course. [Edited for readability and brevity.]
My understanding is that MiniKF is a one-node Kubernetes instance. Does it scale on an on-prem cluster?
Correct, the MiniKF packaged distribution and Arrikto’s Kubeflow as a Service are single VM deployments of Kubeflow. If you are looking for a scalable Kubeflow deployment, we recommend checking out the Enterprise Kubeflow distribution.
I’m wondering if it is possible to train ARIMA models in MiniKF?
If Kubeflow supports ARIMA models, then MiniKF will as well.
Are there any requirements for the docker image used for Katib jobs?
Yes, you learn more under “Running an Experiment” in the official Kubeflow documentation.
Is it possible to access the files, e.g. submission.csv, on the web UI?
Yes. For example, here is the submission.csv viewed from the web UI.