Labelled and unlabelled data use in single model

Shrikant

Read CSV is for initial data. is labelled 

Read CSV (2) is for _new.csv file for prediction is unlabelled.

Inside cross validation:

Is this approach correct?

ceaperez

Shrikant,

Because the second CSV file is used to validate you model, both datasets need to have the same structure, data types and roles.

if you need to change or asign new roles you can use the Set Role operator.

Best, 

Cesar

Shrikant

second file has same structure except it doesn't have one column which is for prediction. In practice first file is the past data and the second one is for prediction. Then that design still correct?

ceaperez

Hi again, 

The design is correct. 

Best