Getting Started with Kubeflow

Quickly get running with your ML Workflow on an existing Kubernetes installation

Who should consider using Kubeflow?

Based on the current functionality you should consider using Kubeflow if:

  • You want to train/serve TensorFlow models in different environments (e.g. local, on prem, and cloud)
  • You want to use Jupyter notebooks to manage TensorFlow training jobs
  • You want to launch training jobs that use resources – such as additional CPUs or GPUs – that aren’t available on your personal computer
  • You want to combine TensorFlow with other processes
    • For example, you may want to use tensorflow/agents to run simulations to generate data for training reinforcement learning models.

This list is based ONLY on current capabilities. We are investing significant resources to expand the functionality and actively soliciting help from companies and individuals interested in contributing (see Contributing).

Set up Kubernetes

This documentation assumes you have a Kubernetes cluster available. If not, set up one of these environments first:

  • Local - there are a several options:
    • Minikube setup
      • Minikube leverages virtualization applications like Virtual Box or VMware Fusion to host the virtual machine and provides a CLI that can be leveraged outside of the VM.
      • Minikube defines a fully baked ISO that contains a minimal operating system and kubernetes already installed.
      • This option may be useful if you are just starting to learn and already have one of the virtualization applications already installed.
    • Microk8s setup
      • The benefits of using Microk8s include:
        • Can be installed on any Linux system as a snap
        • Strong isolation and update semantics - your single-node cluster will be updated within a short period after upstream Kubernetes releases.
        • GPU pass through built in - e.g. microk8s.enable gpu
      • If you are not on a Linux machine, or you want to use Kubeflow in a confined environment, then use Multipass to launch a virtual machine. Benefits include:
        • Ubuntu cloud-images already integrated.
        • Lightweight hypervisor using native operating system mechanisms (e.g. Hypervisor Framework on MacOS, Hyper-V on Windows 10, QEMU/KVM for linux)
        • Eliminates the need to install a separate virtualization application.
        • You can use cloud-init to customize the VM (as you might in a cloud)
  • Cloud:

For more general information on setting up a Kubernetes cluster please refer to Kubernetes Setup. If you want to use GPUs, be sure to follow the Kubernetes instructions for enabling GPUs.

Kubeflow quick start


  • ksonnet version 0.13.1 or later.
  • Kubernetes 1.8 or later
  • kubectl

Download, set up, and deploy (If you prefer to work from source code, feel free to skip step 1):

  1. Run the following commands to download

    mkdir ${KUBEFLOW_SRC}
    cd ${KUBEFLOW_SRC}
    export KUBEFLOW_TAG=v0.4.1
    curl${KUBEFLOW_TAG}/scripts/ | bash
    • KUBEFLOW_SRC a directory where you want to download the source to
    • KUBEFLOW_TAG a tag corresponding to the version to check out, such as master for the latest code.
    • Note you can also just clone the repository using git.
  2. Run the following commands to setup and deploy Kubeflow:

    ${KUBEFLOW_SRC}/scripts/ init ${KFAPP} --platform none
    cd ${KFAPP}
    ${KUBEFLOW_SRC}/scripts/ generate k8s
    ${KUBEFLOW_SRC}/scripts/ apply k8s
    • ${KFAPP} the name for the kubeflow deployment (shouldn’t be a path). A directory with the name will be created under pwd when you run init, and that is where kubeflow configurations will be stored.
      • The ksonnet app will be created in the directory ${KFAPP}/ks_app
    • (optional) For GPU support, make sure your cluster is in a zone that has GPUs. To set the zone explicitly, append --zone ${ZONE} to the init command.

Important: The commands above will enable collection of anonymous user data to help us improve Kubeflow; for more information including instructions for explicitly disabling it please refer to the usage reporting guide.


For detailed troubleshooting instructions, please refer to the troubleshooting guide.