Overview of Retraining Models

Using SAS Model Manager, you can retrain models to respond to data and market changes. Retraining models enables you to update models and to improve model performance. When you define a model retrain task, you can select multiple models to be retrained at the same time. The definition of the model retrain task includes the destination version and training data source. The destination version is an existing version or new version that is associated with the selected project and that stores the retrained model information. The training data source contains new data for retraining the selected models.
The model retrain task enables you to specify a location to store comparison reports and retrain results. When you select the models to include in the comparison report, you can use the training data source or select a different data source to compare the performance of the new models. The model retrain task also enables you to specify the report options, including the name, format, and style of the comparison report. When you define e-mail notifications for a model retrain task, they are sent after you execute a model retrain task.
By default, the champion model for the selected project is retrained if a model is not selected. After you execute a model retrain task, if the Register new trained model option was selected, SAS Model Manager registers the new models to the destination version. The comparison report is stored in the Model Retrain folder, as well as in the report folder on the SAS Workspace Server that was specified when the model retrain task was defined.
Note: Only models that are created by using SAS Enterprise Miner, R models, or SAS/STAT linear models can be retrained.
In this tutorial, you perform the following tasks:
  • define a model retrain task
  • execute the model retrain task
  • view the new retrained models and comparison report