is anyone using null_count in elementary.column_anomalies test?

I've seen your configuration in the AI channel, I don't think it works this way. I believe you should start with something like this:

models:
  - name: my_model
    config:
      elementary:
        timestamp_column: "updated_at"
    columns:
      - name: my_column
        tests:
          - elementary.column_anomalies:
              column_anomalies:
                - null_count
              tags: ["elementary"]

Thanks, is the timestamp column necessary as the table does not have that for now?

It is not strictly necessary, but recommended. Check the documentation here: https://docs.elementary-data.com/data-tests/anomaly-detection-configuration/timestamp-column It describes among other things how the test behaves when a timestamp column is not present. When you say that you don't have a timestamp column, do you mean that you don't have any kind of date or datetime column in the table? Or do you mean that you don't specifically have a column of TIMESTAMP data type? Because I think the latter is not strictly needed, I believe you can amend this easily in the test

tests: - elementary.column_anomalies: column_anomalies: - null_count tags: ["elementary"] its having issue with - elementary.column_anomalies: dict

excellent!