MLOps

- Overview

MLOps, or Machine Learning Operations, is a set of practices that help manage the machine learning (ML) lifecycle. MLOps is a combination of the terms "machine learning" and "DevOps", a continuous delivery practice in the software field. 

MLOps helps ensure that ML models are developed, tested, and deployed in a consistent and reliable way. It involves tasks such as: Experiment tracking, Model deployment, Model monitoring, and Model retraining. 

MLOps can help with:

• Efficiency: MLOps automates and simplifies ML workflows and deployments.
• Collaboration: MLOps enables seamless collaboration across teams.
• Time to production: MLOps can help reduce the time it takes to get ML models into production.
• Reproducible results: MLOps can help ensure that results are reproducible.  

One technique used in MLOps is shadow deployment, which involves deploying a new version of a model alongside the current production model. This allows teams to evaluate the new model's performance without disrupting the live system. 

MLOps are a set of practices that automate and simplify ML workflows and deployments. ML and artificial intelligence (AI) are core capabilities that you can implement to solve complex real-world problems and deliver value to your customers. 

While MLOps began as a set of best practices, it is slowly evolving into an independent approach to ML lifecycle management. MLOps applies to the entire lifecycle - from integration with model generation (software development lifecycle, continuous integration/continuous delivery), orchestration and deployment, to operational health, diagnostics, governance and business metrics.

Please refer to the following for more information:

• Wikipedia: MLOps