Your submission was sent successfully! Close

Thank you for signing up for our newsletter!
In these regular emails you will find the latest updates from Canonical and upcoming events where you can meet our team.Close

Thank you for contacting our team. We will be in touch shortly.Close

  1. Blog
  2. Article

Rui Vasconcelos
on 24 August 2020

New release, increased capabilities.

After 6 months since the release of 1.0, Kubeflow releases a new version with increased capabilities.

This new version has focused on improving ML Workflow Productivity, Isolation and Security, and GitOps. Here is a list of the enhanced features:

  1. Fairing and Kale (Kubeflow Automated pipeLines Engine) for end-to-end workflows.
  2. Katib hyperparameter tuning:
    1. New frameworks & algorithms (goptuna with CMA-ES, DARTS, chocolate, hyperopt, skopt).
    2. Flexible config & tuning options (new python SDK, experiments with undefined goal, new UI, new resume policy).
  3. Install and operations to support GitOps, using blueprints and kpt primitives.
  4. Isolation and security: multi-user Kubeflow pipelines, CVE scanning, and support for Google’s Private GKE and Anthos.
  5. MXNet and XGBoost distributed training operators, simplify training on multiple nodes, and speeds model creation.
  6. Seldon Core 1.1

What’s next:

  • Read the official release blog post here.
  • Suggest new features for version 1.2 on this thread!

Related posts

Andreea Munteanu
17 March 2021

Kubeflow operations guide

Ubuntu Article

Operating MLOps stacks alike Kubeflow in an increasingly multi-cloud world will be a key topic as this market and Kubernetes adoption grow. Kubeflow operations webinar To discuss this topic, Canonical is holding a live webinar next week, on 23rd of March, 5PM UTC. Besides the key points listed below, the webinar will also have a ...

Rui Vasconcelos
8 March 2021

Latest community videos

Ubuntu Article

MLOps community jewels The MLOps community continues to grow and gift us with great content and discussions around the topic! Here are a couple of interesting discussions – a long one (1h) about Kubeflow, feature stores, and other platforms in the MLOps space, and a short one (3 min) on how to manage dependencies: Sneak ...

Andreea Munteanu
22 February 2021

Still figuring out what is Kubeflow?

Kubeflow Article

Kubeflow has become quite popular in the MLOps community as the tool that enables data science teams to automate their workflows from data preprocessing to model deployment on Kubernetes. However, with it’s made of many pieces, and while it keeps evolving, how can you effectively start using? Learn Kubeflow from online courses Started by ...