Accelerate your Machine Learning model development with MiniKF on AWS

preface: MiniKF is a turnkey solution, for model development, testing, and deployment. Now, you can deploy MiniKF to AWS and enjoy powerful GitOps for ML just like on Google Cloud Platform!

A Machine Learning project usually consists of multiple, interconnected steps: data acquisition, data processing, data modelling, fine-tuning, testing etc. Each of these steps can be a separate process, running at its own cadence, with clearly defined inputs and outputs.

Kubeflow is here to assist you during the complete lifecycle of your ML project: experiment and metadata tracking, framework-agnostic training, Jupyter servers, model serving and many more.

But how can we deploy a Kubeflow instance to get started with it? I have good news: MiniKF. Just like minikube_, _MiniKF provides a single-node Kubernetes cluster with Kubeflow deployed on top, without any hassle. And now, we are excited to see the leading Kubeflow distribution on the AWS Marketplace! Let’s see how we can install it and run a few examples.

Kubeflow

Kubeflow is an open-source project, dedicated to making deployments of ML projects simpler, portable and scalable. From the documentation:

The Kubeflow project is dedicated to making deployments of machine learning (ML) workflows on Kubernetes simple, portable and scalable. Our goal is not to recreate other services, but to provide a straightforward way to deploy best-of-breed open-source systems for ML to diverse infrastructures. Anywhere you are running Kubernetes, you should be able to run Kubeflow.

But how do we get started? Do we need a Kubernetes cluster? Should we deploy the whole thing ourselves? I mean, have you seen the Kubeflow’s manifest repo? Don’t panic; MiniKF is here to lift that burden off you.

Data Scientists have been able to easily and quickly deploy MiniKF on Google Cloud, and their laptop via Vagrant, since November 2019. Today, I’m excited to see that MiniKF is available on the AWS Marketplace!

#aws #programming #artificial-intelligence #kubeflow #machine-learning