Help! Aggregating daily values by sum doubles the values

Hi community,

I have a dataset with daily date in a column and another column with accompanying values as decimals. When I try to plot the values per day I choose to aggregate daily values by “sum”. But when I do that, the values are doubled every day. I know the values are doubled because looking at, e.g., April 18 2023 in the dataset the value is 7 but when I visualize it “analysis” the value for that date is 14.

Why is that and how do I fix this?

Why I am aggregating by sum is because I want to be able to aggregate the date by week and months so I want the default aggregation of the values to be the sum.

Thanks in advance!

Hi @martiniukas - Welcome to AWS QuickSight Community .Can you please share some sample data and screenshot ( ensure do not display the PII or PCI information). This will help in understanding the issue better and guide you the solution.

Regards - San

1 Like

Hi @martiniukas - Thanks for the details. I try to replicate the issue but i am getting the desired result. see the screen shot.

Can you create a new visual and only keep the table data see whether you are getting right outcome or not.

Regards - San

Hi @martiniukas ,

I think you have duplicate rows in your dataset.

2 Likes

Hi,

@nshah-quicksight I think so too. When I aggregate in “count” it gives the value 2. However, looking in the dataset and at the table, I should be seeing the duplicate values shouldn’t I? I think that is strange that I don’t. I have also verified in SageMaker, where I am producing the data, that there aren’t any duplicates in the dataframe before I upload it to S3. Could there be issue when uploading to S3 and then to Quicksight that the data gets duplicated hiddenly?

@martiniukas There shouldn’t be any duplications from s3 uploading.

Is this issue still persistent?