More of a Question

Hi,

Apologies as this not an 'issue' but just had a question regarding your approach to the Kaggle Lending Data competition. I see that in the code you downloaded ~11 columns however the raw data has far more columns (features) Why did you select those particular features instead of others?

Thanks and amazing work!!
