Hi All,
I have data coming from a narrow Timestream database, the timestamps for the different series are not synchronised - as they are only sent when the data has changed more than a set amount.
I need to sum two of the series “Series A” and “Series B”.
To do this I have been trying to use the “Add field from calculation” → “Reduce Row” → “Last *” function as I believed this would fill in the null values with the last reading.
However it doesn’t seem to do anything to my data.
Hi, AFAIK Add field from calculation will take default 0 if the field is empty (you can see that in your Total column), so if your case is just sum A and B in each timestamp - it’s just it.
UNLESS what you want is to have two numbers as a result - sum of all values of series A and sum of all values of series B (which I’ve thought about just now), you can then use Reduce transformation (something like that) that will reduce each column with a certain calculation (use total for your case).
Sorry for the slow reply, that seems to match what I have found. The closest effect I have found so far is (mis)using a centred mean with a small window, however this will only fill in short gaps.
Is there any way to fill downwards using the last value in a transformation or another way to calculate the sum using the last value of each column?
