My data set is an N x N matrix and it represents a scalar field.
The matrix representation is done by defining my own scalar field class. This way, I can avoid double
