Data Preparation

The input Execute file structure is the same as the Train one, except for the number of concat codes and descriptions, which could differ in the Execute phase. The logical distinction between level and concat fields is that the levels are compared one by one (Train level 0 against Execute level 0; Train level 1 against Execute level 1, etc.), while all concat fields in each phase are first concatenated in a single long field and then compared (Train concat against Execute concat).

The same data preparation for the Train phase also applies to the Execute phase.