The Azure Machine Learning Practice guides are short, resource loaded guides designed to help you (or your data science practice) quickly get started and to successfully deliver machine learning projects utilizing Azure Machine Learning.
The guides provide best practice and process guidance, along with templates and checklists, and where appropriate refers to samples publicly available on GitHub when sample code or solutions are helpful.
The Delivery Guide takes a high level view of the overall process of delivering a machine learning solution, going from the initial customer conversation, envisioning sessions, architecture design sessions (requirements gathering and clarification, whiteboarding), PoC/Pilot delivery, operationalization and post-deployment considerations. This guide is about 10-15 pages in length and includes supporting checklists and templates to help you get started quickly and stay organize throughout project delivery.
The PoC and Pilot Guide is a comprehensive and authoritative process-centric guide for planning, designing and executing machine learning PoC’s and pilots in Azure and utilizing Azure Machine Learning. This guide covers the important topics to know before talking to the customer (inclusive of topics around performance, availability, scalability, security, maintainability, extensibility, cost, tools and key services). The guide is about 10-15 pages in length.
Templates and checklists are provided for each phase of the PoC or pilot delivery, centered around specific machine learning horizontal scenarios (computer vision, text analytics and NLP, recommendation, forecasting, anomaly detection, classification and regression and reinforcement learning).