Good day. I'm assuming you're trying to submit the prediction on the test data in your you got from your train_test_split. You shouldn't. That's more like a validation data set to add an extra layer of testing. Instead, after you are comfortable with your models performance on this validation data set, make predictions on the test.csv data and submit that.
Good day. I'm assuming you're trying to submit the prediction on the test data in your you got from your train_test_split. You shouldn't. That's more like a validation data set to add an extra layer of testing. Instead, after you are comfortable with your models performance on this validation data set, make predictions on the test.csv data and submit that.
Yes I used:
Submission = pd.DataFrame({'Applicant_ID': test['Applicant_ID'], 'default_status': y_pred})
Alright then. Just make sure your y_pred is based on the data in test.csv and you're good to go.
And also be sure you are using the second Predict proba too
OK. Thank you all.