Kubeflow 1.0 has been released today and Canonical would like to take this opportunity to congratulate the community for their hard work and leadership (link).
What is Kubeflow?
Kubeflow is an open source artificial intelligence / machine learning (AI/ML) tool that helps improve deployment, portability and management of AI/ML models. This tool allows users to quickly create, train and tune neural networks within Kubernetes for dynamic resource provisioning. It helps data scientists and AI/ML developers who want to easily setup a project or pipeline by unifying all the required tools under one installation. Kubeflow works well with TensorFlow and other modern AI/ML frameworks such as PyTorch, MXNet and Chainer allowing users to enhance their existing code and setup.
Read more about it here.
In order to support the community’s efforts and accelerate adoption, Canonical is committed to natively include Kubeflow in both of it’s Kubernetes solutions; MicroK8s and Charmed Kubernetes. For instructions on how to get started follow this guide here.
To further accelerate the adoption of Kubeflow in production environments Canonical will provide it alongside Charmed Istio and Charmed Seldon Core. This will continue in Canonical’s mission to grow the adoption of open-source tools and democratisation of AI/ML.
Accelerating Kubeflow adoption in production
This 1.0 release is a milestone that signals Kubeflow’s readiness to underpin production AI/ML environments for a wide array of industries. In order to further accelerate its adoption, Canonical has partnered with key players in the realm of hardware manufacturing as well as enterprise professional services, for example:
In the realm of cloud hardware manufacturers, Raju Penumatcha, CPO & SVP of Supermicro told us: “Supermicro is very excited about the partnership with Canonical to deliver the best-of-class AI workflow orchestration on the state-of-the-art Supermicro systems. Customers will substantially benefit from the ease of use, flexibility, and reliability delivered by the Kubeflow on Supermicro solution to run scalable, operational AI applications.”
In the area of AI/ML professional services, Al Kari, CEO of Manceps told us, “We are really excited about the work we’re doing with Canonical to deliver production-grade machine learning scale-out architectures built on Kubernetes with Kubeflow.”
The future of Kubeflow
This 1.0 release means reliability, stability and quality. Further functionality will soon be added to the 1.0 release but the core select ones of today showcase the potential this has for the AI community. Canonical is excited to continue supporting Kubeflow and accelerate its adoption worldwide.
For any help with AI/ML, please contact us
With Charmed Kubeflow, deployment and operations of Kubeflow are easy for any scenario.
Charmed Kubeflow is a collection of Python operators that define integration of the apps inside Kubeflow, like katib or pipelines-ui.
Use Kubeflow on-prem, desktop, edge, public cloud and multi-cloud.
Kubeflow makes deployments of Machine Learning workflows on Kubernetes simple, portable and scalable.
Kubeflow is the machine learning toolkit for Kubernetes. It extends Kubernetes ability to run independent and configurable steps, with machine learning specific frameworks and libraries.
The Kubeflow project is dedicated to making deployments of machine learning workflows on Kubernetes simple,
portable and scalable.
You can install Kubeflow on your workstation, local server or public cloud VM. It is easy to install with MicroK8s on any of these environments and can be scaled to high-availability.