Machine Learning on Google Cloud with H20
Details
Please sign in at the front door.
Kubeflow is an extra layer of abstraction built on top of Kubernetes that allows easy deployment of Machine Learning jobs and tasks in a convenient and highly-scalable environment. Kubeflow Pipelines adds additional functionality and ease of use to the existing Kubeflow infrastructure by creating an environment for developing and managing full machine learning workflows.
Driverless AI is H2O.ai’s enterprisesolution that features automatic machine learning. It performs many of the standard tasks performed by data scientists, including feature engineering, hyperparameter tuning, and model stacking/ensembling.
Finally, Driverless AI is capable of producing a lightweight
and easily deployable final model pipeline (MOJO) to place your model into production. Data science is not only about training models. There are multiple steps, ranging from obtaining your data, to training a good model, to putting that model into production.
During this session we will be talking about the integration between Kubeflow, Kubeflow Pipelines, and H2O.ai’s Driverless AI. This will include a discussion of how to deploy static instances of Driverless AI to Kubeflow, leveraging Kubernetes and Kubeflow to manage compute resources like GPUs. Additionally, we will talk about how to use Kubeflow Pipelines to manage and deploy full machine learning workflows, from data ETL to model deployment.
Schedule:
6:00-6:30pm Networking
6:30-7:30pm Talk
7:30-8:00pm Networking
Nicholas Png is a Partnerships Software Engineer at H2O.ai. Prior to working at H2O, he worked as a Quality Assurance Software Engineer, developing software automation testing. Nicholas holds a degree in Mechanical Engineering, and has experience working with customers across multiple industries, identifying common problems, and designing robust, automated solutions.
About Metis
Metis (thisismetis.com) accelerates careers in data science by providing full-time immersive bootcamps, evening part-time professional development courses, online resources, and corporate programs based in Seattle, New York, Chicago, and San Francisco.
Brought to you by Kaplan, Metis focuses primarily on Python, machine learning, data visualization, deep learning, big data processing, statistical foundations, and more. Students and alumni of the bootcamp program receive continuous support from our career advisors, empowering them to pursue a successful career in the fast-growing field of data science.
Learn more about us at
Join our Metis Community Slack channel!
Apply here:
http://bit.ly/MetisCommunitySlack
Metis Code of Conduct
Metis is dedicated to providing a harassment-free experience for everyone, regardless of gender identity, age, sexual orientation, disability, physical appearance, body size, race, or religion (or lack thereof).
We do not tolerate harassment of students, staff, or visitors in any form. Sexual language and imagery is not appropriate for any event including talks, workshops, parties, and other online media. Individuals and groups that do not abide by these rules will be asked to leave and, if necessary, prohibited from future events.
If you have any questions or you're made to feel uncomfortable by anyone on our campus or at one of our offsite events, please let one of the staff members know right away. The matter will be taken seriously and promptly addressed.