A step-by-step guide on installing and configuring each of the kubeflow components on your local machine.
Kubeflow is an open source platform developed by google to contain the machine learning model development life cycle. Kubeflow is made up of a set of tools that address each of the stages which compound the machine learning life cycle, such as: data exploration, feature engineering, feature transformation, model experimentation, model training, model evaluation, model tuning, model serving and model versioning. The main attribute of kubeflow is that it is designed to work on top of kubernetes, that is, kubeflow takes advantage of the benefits that a kubernetescluster provides such as container orchestration and auto-scaling.
While kubernetes is an standard technology for container orchestration, it can be a complex and time-consuming process for data scientists or machine learning engineers to configure and coordinate each stage of the machine learning life cycle directly in a kubernetescluster. Thus kubeflow comes as the platform that provides the tools to configure, develop, automate and deploy each stage of the machine learning life cycle in a kubernetescluster, avoiding data scientists or machine learning engineers from having to spend time on configure and apply changes directly in the kubernetescluster.
In this blog we are going to see how to install and configure kubeflow on your local machine in order to be able to start using kubeflow locally without the need for a cloud provider. Therefore, this blog will be divided into:
#docker #kubeflow #kubernetes