What is required before starting a scoring run?

Before starting a scoring run in the context of using Adobe Experience Platform, it is essential to ensure that a training run has been completed. The training run is a critical step in the machine learning process, where the model is developed and refined based on historical data. This process involves using the data to teach the algorithm how to make predictions or score new data effectively. Only after this is done can the model be applied to new data during the scoring run.

Completing a training run ensures that the model is adequately trained and has learned the relationships within the data, making it capable of delivering accurate and meaningful insights when scoring new datasets. This foundational step helps avoid using untested or ineffective models, which could lead to incorrect predictions or scoring outcomes.

In a typical workflow, other steps like data ingestion or model review would follow, but they are not prerequisites for initiating a scoring run. Therefore, having a completed training run is the key requirement to ensure that the model is ready to score new data effectively.