Redshift Avg 不返回小数

Question

我不知道如何在平均值计算中 return 小数。

我有：

select
    r_i_flag,
    mr_age_bucket,
    count(request_id) as sum_requests,
    avg(age_of_mr_days) as avg_age_of_mr_days,
    median(age_of_mr_days) as median_age_of_mr_days
from
    together
group by
    1,
    2
order by
    1 desc,
    2

我试过使用 to_number 将 mr_age_bucket 转换为数字，尝试将 avg/median 转换为数字，小数，但我得到的只是四舍五入的数字：

r_i_flag	mr_age_bucket	sum_requests	avg_age_of_mr_days	median_age_of_mr_days
零售业	<7 天	29	1	1
零售业	>7 天	107	46	30
机构	<7 天	1	5	5
机构	>7 天	13	45	29

我觉得这是一件非常愚蠢的事情，但我想不通。我如何 return 一个非四舍五入的数字？谢谢

Answer 1

问题是传递给 AVG 的数据类型，只需将其除以 1.0 即可。投射到 DOUBLE PRECISION 或 REAL 也可以

select
    r_i_flag,
    mr_age_bucket,
    count(request_id) as sum_requests,
    avg(age_of_mr_days / 1.0) as avg_age_of_mr_days,
    avg(age_of_mr_days::double precision),
    avg(age_of_mr_days::real),
    median(age_of_mr_days) as median_age_of_mr_days

from
    together
group by
    1,
    2
order by
    1 desc,
    2