Hello, fellow competitors.There are several ways to decide which features will be in your model, and one of my favorites is recursive elimination.For this compe... See more

This was a great competition thanks @Zindi and @PredictiveInsights. It has reminded everyone(some more than others) on the importance of cross-validation and ov... See more

Iam just confused on how to Identify Missing at Random and Not Missing at Random,Could someone who have a clear view or who seem to grasp those concepts explain... See more

If I drop NaN vales in the dataset, which means that some IDs are not present in the submission file, does that mean I won't be scored when I submit?

Please i don't whats happening each time i make a submission, it keeps on displaying scoring and never displaying anything

Has anyone else witnessed a lag in getting scores? Is it taking a lot of time to get the score after submitting?

Join me next Saturday the 26th of August as we go through this Tabular Classification project together. We will go through an EDA driven approach to create an e... See more

What is Schoolquintile?

My cv score has no relation at all with the lb score, sometimes some increase in cv leads to lb advancement but some times it is the opposite, I am using 5 fold... See more

Does removing outliers necessarily produce errors when scoring on unseen data during submission?

I don't understand this part where all the unemployed individuals have tenure values, how can you be unemployed and have been on a labour's activity

on the training set, if a person was surveyed on march 2023, how did you get the target in this case?according to the description, you must've asked him a year... See more

Hello everyone, i wanted to inquire is it okay to use python for the project or its specifically R

I have tested some aggregation features, but the most useful ones were median aggregations,they really did well in lb, they are better than mean aggregations fo... See more

Hello,Is there a place to read the dataset schema to understand what each column means in other to make educative hypotheses? ( I may have missed it. please dir... See more

Good day,Does anyone know what the variables tenure and round are referring to in this dataset?Thanks in advance

predict_proba(X_test)[:, 1] is more adequat!

Hello, thanks for a fun competition. Can you tell us anything more about the `status` variable, for instance, does the survey contain a free-text entry here, or... See more

I think some Hackathon should be more based on report based than leaderboard submissions score.Let everyone get hold of the dataset, find insight and communicat... See more

I will love to thank Zindi and predictive insight for this great opportunity, but I will love to make some point about the prize that is only valid to South Afr... See more