You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When aggregating data, we can already compute percentiles for each result of the transform. However, If we need to compute a percentile aggregation on multiple lines of the transform-aggregated data, it's impossible because we need the data distribution to compute a percentile. Histograms are perfect for that.
Thanks for the feedback. I agree the histogram field type would be the way to go as output of a histogram aggregation in a transform, because for ordinary mappings we need to know the buckets in advance.
Hello,
Given the Transform API documentation, we can not yet create a transform using Histogram as a
pivot.aggregations
.This would be quite useful to store a Histogram value using histogram field type as a result of a transform.
Why?
When aggregating data, we can already compute percentiles for each result of the transform. However, If we need to compute a percentile aggregation on multiple lines of the transform-aggregated data, it's impossible because we need the data distribution to compute a percentile. Histograms are perfect for that.
Adding Histogram Transform aggregation would allow this using the value of the histogram field type.
Thanks
The text was updated successfully, but these errors were encountered: