Aggregate Transformations
Watch this short demo to see how to aggregate transformations with Datameer
AGGREGATE
Aggregates or aggregations are extremely valuable data transformation operations. Typically, an enterprise data warehouse requires writing custom SQL code that can get pretty complex and lengthy—resulting in lengthy data engineering cycles.
Datameer, on the other hand, provides a friendly, easy end-user interface to implement the aggregate transforms using a no-code approach.
- Here I have a LOAN_DATASET that has some banking loan data, and as you can see from the bottom, it has the MEMBER_ID of the person who took the loan, the loan amount, the interest rate of the loan, and it also has the geolocation of the person who took the loan.
- What if I wanted to aggregate this data by geolocation and product measures like an average interest rate for location or loan amount totals over the location. To do that, I am going to create a new transformation. I have joins, pivots, filters, etc.
- I'm going to choose the aggregate. And it opens the aggregate dialog, a highly intuitive wizard-led interface that doesn't require you to write a single line of SQL code.
- I'm going to go ahead and add a couple of my geographic dimensions, which are my LAT and LONG.
- And I am going to add a few measures, including an interest rate average. You do have other options, as you can see here. And also a total on my loan amount, or sum. Go ahead and check the count, which gives me the total count by location.
- You can validate the results and look at your data interactively. And once you do that, hit Apply the transformation. It adds a transform to a new aggregate view, to the lineage. If required, you can add other transforms to this flow, like joins and filters. And ultimately publish this to Snowflake.