Reload Dataset

In Max, when a dataset is first built, Max reviews each dimensional column to identify the unique values found in those columns. These dimensional values are used to ground natural language questions, helping to address typos and inconsistencies between what the user typed and how the data is stored in the database.

When the underlying data is updated, users should reload the dataset to ensure it contains the most current valid values. This can be done using the Reload Dataset option, available in the right menu for each dataset.

Reload Dataset can be done programmatically through the AnswerRocket SDK. This allows your ETL pipeline to keep Max in sync after the pipeline has run successfully.