It would be useful to add new wrappers (i.e. subclasses of [`Approximation`](https://github.com/mathLab/EZyRB/blob/48af7bd1f02ac186be88dbcbcead53ff4eb88830/ezyrb/approximation.py) for interpolators in order to improve usability from EZyRB. [Here](https://github.com/mathLab/EZyRB/blob/48af7bd1f02ac186be88dbcbcead53ff4eb88830/ezyrb/rbf.py) you can find an example, the EZyRB wrapper for `RBF`. Each wrapper should contain the methods `fit` and `predict` to allow training and using the interpolator. Some resources about `scipy.interpolate`: + [GitHub](https://github.com/scipy/scipy/tree/master/scipy/interpolate); + [API reference](https://docs.scipy.org/doc/scipy/reference/reference/interpolate.html#module-scipy.interpolate); + [SciPy Tutorial](https://docs.scipy.org/doc/scipy/reference/tutorial/interpolate.html)