https://www.kaggle.com/code/danielbruintjies/absa-client-activity-forcasting-challenge/notebookSolution: Ensemble of 4 Multivariate Multistep LSTM with UserID E... See more

Hi @zindi @amyflorida626If you don't select any subs, then the platform will enter your best 2 from public, correct?I selected just one sub, assuming the platfo... See more

@ZindiI am struggling to submit, every time when I try to submit it gives me an error and I do not know what it means.Apolologies if this is trivial, its my fi... See more

Hi! can anyone guide me here I am lost here. How did you create a target columns?Any tips for doing so?

I attepted to start off with a linear model and continuously adjust the model. That is start with a modle that guesses the same event for all inputs and continu... See more

Hii @amy, I noticed that there are rows with the same eventdatetime and event duplicated for the same userid. If we are under the assumption that a row represen... See more

@zindiSeems there is lots of missing data ... big problem.If there are no target events for a given user in a given timeslot, then it would make sense to assume... See more

Does the order of the instances have an impact on the score?

@zindiIt seeems userids have multiple emails or countries or roles ...Which user are we predicting for in test?

If my score is just slightly higher than 0, then my position should be above those with a zero score :-)

I noticed only seven(7) variable definitions in the training dataset even though eight(8) are specified on the comp. page. Perhaps it's a spoof?The training dat... See more

correct me if i'm wrong.although the decription explains data as following 'The training set is the first six weeks of data, and the seventh and final week will... See more

Hey there. I joined the competition, and it says it starts after 11 days. Last time I checked it said it starts after a day. Has it started and if it has, where... See more

public data such as FX rates, inflation etc.