Are there any cases in the test data where there is a gap between the last timestamp provided for a given process and the start of the target time period to predict? In other words, if a process only has a pre-rinse and caustic phase, the target time period containing the final rinse occurs immediately after the caustic phase (i.e. there was no intermediate or acid phase for that process)?
For the test data, only the final_rinse phase has been removed. Any other phases present in the original dataset are still there.
This, of course, was not accurate. The test set is subdivided as described in the problem description:
Hi all, we just made an announcement releasing “recipes” that specify the phases you can expect (for the most part) for each process. Find the announcement here (you must be logged in to see this link):
https://www.drivendata.org/competitions/56/predict-cleaning-time-series/announcements/