Hello~!
What is the difference between FDW and Historical rows?
I understand Feature Derivation Window is used for generating rolling derived features.
What is the specific definition of Historical rows on "Make Predictions" tab?
I thought these were almost same but sometimes the numbers were different each other.
case 1) FDW = Historical rows
case 2) FDW ‡ Historical rows
Could you help me to understand it?
Thank you.
Hi!
Historical rows under the prediction tab mean how many rows of data satisfy criteria to create derivation features. for each row you will have relative amount of forecast rows predictions. Probably, in your example, Historical rows + Feature derivation window should equal size of your backtest.