A Prototype for Machine Learning Model Deployment In Cloud Environment

Abstract

Training and developing a Machine Learning (ML) model is a difficult task, and then after successfully creating a working model, deploying and distribution is an added feature. In most instances, those models are never deployed. To help with this, we present a prototype, a SaaS platform to allow users to dynamically deploy their machine learning models to the cloud and host them so that the user has complete control over the visibility and accessibility.This delivery and deployment model provides lower upfront cost, timely updates, and a dedicated work/host environment. The platform's sole purpose revolves around the idea of a sharable deployable and ready to use Machine Learning Model. It takes advantage of the Continuous Integration and Continuous Delivery archetype facilitated by Kubernetes to dynamically provide custom and updated with the latest libraries docker environment. © 2023 IEEE.