data splitting from credit risk modeling
I don't understand this line of code
train_test_split(loan_data.drop('good_bad', axis = 1), loan_data['good_bad'])
why do we drop [good_bad] and then expect it in loan_data['good_bad'] as a result. I would like to have more clarification about. I failed to understand it for multiple time
1 answers ( 0 marked as helpful)
Hi!
Our target var is 'good_bad', thus it has to be separeted from all other vars which are the inputs.
In the parentheses the first item (loan_data.drop('good_bad', axis = 1) refers to inputs and second one to the target var itself (loan_data['good_bad']).