In a regression model with one input variable, consider the case of using linear splines with 11 knots. Looking at the spline model as a linear regression model with the input variables corresponding to (some) spline basis functions. if you run an exhaustive best subset selection in order to select the best subset, how many different models would you have to consider