--- data generating process simulator ----
This function simulates a data generating process, by drawing random samples according to a distribution of choice from
- standard gaussian
- gaussian with fat tails
- gamma type distributions.
In addition, to the standard parameters of a data simulator, available parameter choices are:
- number of non-linear features
- number of irrelevant features
- highest polynomial degree
It can be used to simulate data for binary classification problems or regression problems.