Counting non identical columns

Could anyone explain how to count how many, or remove  non-identical entries there are in a set of columns?

e.g. In this table, in lines 1- 4 column A and B are identical, but lines 5 - 9, column A is the same in each but column B has a different value in line 9 than in lines 6-8.

I want to find a way to count lines 1-5 as one entry and lines 5 - 9 as two entries (i.e. same code in all five lines but different date in one of them.) so the number of unique entries is 3.