Closed abgoswam closed 7 years ago
thanks! can you address the comments?
The additional amount of events we have to load depends on the experimental unit duration. For Track Revenue it’s 20h (or so).
Maybe experimental_unit_duration = 1 An interpret the number as days?
Oh, then just remove.
This PR addresses issues related to mismatch of data range.
With this fix, we now see missing events 0 for an example run with TR dataset where previously we were seeing ~1K missing events