MLOps with Azure Machine Learning
Accelerating the process of building, training, and deploying models at scale
Enterprise organizations in nearly every industry are increasingly making machine learning (ML) initiatives a priority. Machine learning models are at the core of many AI use cases from predictive maintenance and recommendations to anomaly detection and dynamic pricing.
Modern machine learning algorithms and frameworks make it increasingly easy to develop models that can make accurate predictions. But even when you build a model that exceeds expectations, putting it into production has its own share of challenges. Over time, even a once successful model might deteriorate, or perhaps you think you can build an even better model to fit the changing needs of the business. The process is never- ending, and it only becomes more complex as your business grows.
Enterprises investing in ML initiatives should consider implementing a machine learning DevOps strategy. Machine learning DevOps (MLOps) is an organizational change that relies on a combination
of people, process, and technology to deliver machine learning solutions in a robust, scalable, reliable, and automated way.
This whitepaper overviews best practices for MLOps with Azure Machine Learning. It covers the technical capabilities of Azure ML and how it supports model reproducibility, validation, deployment, and monitoring. But just as importantly, we’ve thought through the key considerations and recommendations for improving processes and making it easier for teams to collaborate. Ultimately, the goal of MLOps–your goal–is to close the gap between development and production and deliver value to customers faster. At Microsoft, our goal is to provide you with the best tools–tools that support your organization’s processes and people wherever you are on your ML journey.