what is

elementary.expression_anomalies

? does it accept the same params as column_anomalies ?

The knowledge sources do not contain any information about an

elementary.expression_anomalies

test. This test name does not appear in the Elementary documentation or community discussions referenced in the available sources. Based on what the knowledge sources do cover, Elementary does not natively support running anomaly detection directly on an arbitrary SQL expression as if it were a column. As confirmed in a community discussion, if you want to monitor a calculated value (e.g., the difference between two timestamp columns), you need to materialize it first as a column in a model or view, then apply Elementary tests to it. [Slack discussion] If you're looking to monitor a SQL expression, here are the supported approaches in Elementary: •

dimensions

parameter – Available on

column_anomalies

,

all_columns_anomalies

, and

dimension_anomalies

. Accepts SQL expressions for grouping your anomaly detection. [Dimensions config] •

where_expression

parameter – Available on all anomaly tests. Accepts a SQL expression to filter the data being tested. [Anomaly params] • Materialize the expression – Create a dbt model or view with the calculated column, then apply

elementary.column_anomalies

to it. [Slack discussion] It's possible you may be thinking of a feature that doesn't exist yet, or perhaps a different test name. I'd recommend checking the official Elementary docs or the Elementary community Slack for the most up-to-date list of supported tests. 👍 0 button 👎 0 button ℹ️ Tip: Mention @elementary.ai in the thread for follow-ups or DM the app for 1-to-1 help.