Forum Discussion
Creating unique key: Ending up with duplicate values after using Groupby function
Hi there,
Im trying to create a unique key using the groupby function. I want to have a row count/sum of a value for every unique key. And it works good until I try to add an additional value to the key (an adress).
I ran a frequency tabel on row count and unique row count and I see that there are key's that occur more than once in the table. When I create a key without the adress it works just fine. I cleaned the adress column using cleaning/trim spaces, refreshed data, double checked my queries, all to no avail.
I've added a screenshot of the frequency table; 1st column is the key, 2nd is unique count, 3rd is the normal count.
Does anybody have experience with this?
Hi Smoupre,
Thnx for the tip.
I think it was a memory problem though.
I made a duplicate query and removed/filtered values that were unique.
The table was empty (as it should be)
Then I refreshed the whole PBI and I got a memory error during refresh (with 32GB Ram..)
After that I shut down PBI and reopend the dashboards after which it was working correctly...
3 Replies
If you are referring to the CHRISTTIAN HUYGENS entry, I see different dates and that seems to be the unique piece that is causing the duplicate entry if I am understanding the problem correctly.
Hi Smoupre,
Thnx for the tip.
I think it was a memory problem though.
I made a duplicate query and removed/filtered values that were unique.
The table was empty (as it should be)
Then I refreshed the whole PBI and I got a memory error during refresh (with 32GB Ram..)
After that I shut down PBI and reopend the dashboards after which it was working correctly...
Cool, glad you got it fixed. I recently had a similar kind of issue with a data model that I built, drove me bananas until I rebuilt it and then it worked perfectly.
