MLOps
MLOps (Machine Learning Operations) is a set of practices and tools used to manage the lifecycle of machine learning models in production environments. It is an extension of DevOps (Development Operations) and aims to bring the same level of automation, collaboration, and agility to machine learning development and deployment.
ML is revolutionizing the world by modernizing industries like healthcare, education, transportation, food, entertainment, and various assembly lines, among others. However, some of the bitter truths of the ML world, when taken to production are:
Deploying machine learning models to production takes much more time than creating them.
In real-world machine learning systems, the actual code that does the machine learning work is only a small part of the overall system. The infrastructure surrounding the code in the production environment is complex and extensive.
Historically, ~85 % of ML models that are built never reach production. Surveys & reports also suggest that only ~60% of projects make it from prototype to production — that too at organizations that have a decent experience with AI. Wikipedia defines MLOps as:
MLOps is the process of taking an experimental Machine Learning model into a production system.
