The identity matrix is a matrix with ones on the leading diagonal, and zeros everywhere else. It's commonly denoted by the letter I if the size is evident from the context.
It can also be denoted by In to refer more specifically to an $n$ by $n$ identity matrix.
It's significant in linear algebra because any matrix multiplied by its identity matrix results in the same as the original.
Implementing in Numpy is trivial and easy to remember.