在 Google Data Studio 或 SQL 中避免数据聚合中的“无穷大”

Question

I am building a report in Data Studio, and the data comes from a BigQuery dataset.

The problem is that some values in the table are Infinity , and when Data Studio aggregates in totals or average, the result is always Infinity.

I need to avoid those values in those aggregates. I am trying to do that in Data Studio, or to replace those values by zero in the SQL query.

Is there a way to do either of those things? Is there another way to deal with this?

Thanks in advance.

Answer 1

Using a Case Statement in your aggregation is likely where you'd like to go. If you want to replace this with 0 then the below logic will work.

SELECT
   SUM(
      CASE
         WHEN col2 = inf THEN 0
         ELSE col2
      END
    ) as aggregated_value
FROM TABLE 1;

If you need to do an average, then 0 isn't a good replacement. Instead use NULL as SQL will ignore NULLS in an average:

SELECT
   AVG(
      CASE
         WHEN col2 = inf THEN NULL
         ELSE col2
      END
    ) as aggregated_value
FROM TABLE 1;

Answer 2

You can use the "is_inf" function in the where clause and remove any bad data records before aggregation

SELECT
   SUM(col2) as aggregated_value
FROM TABLE 1
WHERE IS_INF(col2) = false