A model describes and explains the relationships that exist between the dataset and the target to allow predictions. One model can contain several model versions, but only one version can be active at a time. The active version is used to do the predictions.
In the detail steps of training a model, as an example, we use the Check Assigned Liquidity Items predictive scenario. For other predictive scenarios, the procedures are similar, but you can use our example and adapt it to other scenarios.