You have created and trained a machine learning model with KNIME Analytics Platform. But how to put it into production so that it can produce predictions? In this course, we will show you how to use KNIME Software to test and deploy a prediction workflow, automate its deployment and enable the subsequent continuous deployments, monitoring, and maintenance.
We will use a credit scoring use case as an example to demonstrate how to deploy a prediction workflow manually, automatically, or continuously, and how to generate predictions via a data app or as a REST service.
In the first session of this course, you will learn how to prepare a prediction workflow for deployment. In the second session, you will be introduced to KNIME Business Hub and will learn how to deploy a prediction workflow as a data app or as a REST service. Next, in the third session, you will learn how to use the Continuous Deployment for Data Science (CDDS) framework to enable continuous deployment on KNIME Business Hub. Finally, in the fourth session, you will learn about monitoring machine learning models, and about the best practices to productionize machine learning models, such as, experiment logging and tracking, performance optimization, AutoML, and XAI.
This is an instructor-led course consisting of four, 75-minutes online sessions run by our KNIME data scientists and solution engineers. Each session has an exercise for you to complete at home and together, we will go through the solution at the start of the following session. The course concludes with a 15 to 30-minute wrap up session.
Session 1: Preparing for Deployment
Session 2: Introduction to KNIME Business Hub
Session 3: Continuous Deployment for Data Science
Session 4: Best Practices when Productionizing Data Science
Session 5: Optional follow-up Q&A (15-30min)