This applied, hands-on course teaches you how to manage models through their useful life cycle. You start by creating a modeling project, and then add and compare models to it so you can identify a champion model. The course uses models that are created using SAS Advanced Analytics capabilities and Python and R languages. The course also shows how to implement procedures that ensure that model governance and oversight approval is being followed by implementing workflow.
You learn how to test a model in the production environment to which it will be deployed. After the model test runs successfully, you learn how to schedule the model to run automatically.
Further, the course shows how to measure and monitor the ongoing performance of model accuracy over time. The performance monitoring process will also be scheduled to run automatically in class.
An optional lesson shows how to register and score four types of SAS Visual Text Analytics models.
The self-study e-learning includes:
- Annotatable course notes in PDF format.
- Virtual Lab time to practice.
Learn how to
- Manage SAS Model Manager data sources.
- Import models into SAS Model Manager.
- Score SAS Model Manager models.
- Create SAS Model Manager performance reports.
- Schedule Model Manager jobs.
IT staff who are involved in data preparation and scoring; modelers who create and test models; and business analysts who are consumers of the model, as well as business analysts or consultants who are responsible for integrating models, business rules, and rule flows into operational processes
Managing Models in SAS Viya
- Managing reports and pages.
- SAS Viya architecture.
- Project setup.
- Import models.
- Model properties.
- Publishing models.
- How to define a CAS publishing destination.
- Scoring deployment.
- Creating a Model Performance report.
- Scheduling a performance job.
- Model retraining (self-study).
- SAS Visual Text Analytics.
- Registering Visual Text Analytics models in the Visual Text Analytics Repository.
- Scoring and exploring concept models.
- Scoring and exploring sentiment models.
- Scoring and exploring topic models.
- Scoring and exploring category models.
- Model repositories.
- How to fit a scoring script for model containerization.
- Prepare an R model and PMML file.
- Calculate fit statistics for an R model.
- Feature contribution index.
- Model usage summary.