In the training data we know the duration/interval of the target time period but this is not provided in the test data. Can we assume that none of the final phase data is truncated in the test set?
The training and test sets come from the same original data, so the final rinse phases present in the data will look similar across both.
And how were these target times defined? In other words, why would a measurement in the final phase have target_time == False?
This is covered in the problem description at the end of the Dataset section:
Here’s the relevant portion:
The target time period is the portion of the final rinse phase when the return caustic and return acid valves have been closed for the last time.
We’ve pre-calculated that for you so that you can just use the flag target_time_period and not determine when the valves have been closed for the last time.
Alright, my apologies. I must have read over that.
Thanks again for clearing this up!