
Building ML pipeline with Kubeflow
Kubeflow is an open source machine learning MLOps platform which makes it easy to deploy and manage ML stack on Kubernetes. In this tutorial, I will demonstrate how to create ML pipeline in Kubeflow. We will train and serve image classification model using MNIST dataset. The goal here is to create pipeline for getting data, pre-processing it and creating the model and finally inference the model. Install Kubeflow Make sure your system have minimum 8vCPU and 16GB RAM....