Our target for this contest is “percent decrease in oil saturation”. What is this? Clearly at least 12 of you know as you already have scores on the leaderboard. While I have some guesses, I would like to hear a clear statement.
Thanks to all.
They way I understand it is “current oil saturation divided by the original oil saturation (AVG_ORIG_OIL_SAT) in the given well (CMPL_FAC_ID) and layer (SAND)”. But I agree the data definition is not very clear. Maybe office hours will shed some light.
- Are all the wells in data set only producers? Or some of them are injectors?
- We have Lin_Dist_Inj_Factor and Lin_Dist_Prod_Factor. How is it possible some of those values have NaNs (empty values)?
- How is it possible that given well/layer combination has missing values for injection 1 (both SGMT_CUM_STM_INJ_1 and FT_DIST_PAT_1) but do have values for injection 2? Isn’t 1 supposed to be the closest one and 2 the second closest?
you can have a score and have no idea that it is oil saturation