We show how nonlinear feature maps project the input data to transformed spaces, where they become linearly separable.
