Write a function linearModelLossLAD which computes the least absolute deviation loss function for a linear model parameterized by $,$

as well as the gradient of the loss. The function should take as its first argument a $1$d$-$array beta of coefficients for the linear model, as its

second argument the design matrix $x$ as a $2d-$array, and as its third argument a $1$d$-$array $y$ of observed outcomes.

Test the function with the values:

$x=np*array(\left[\begin{array}{cccc}1& 0& -1& 2\end{array}\right]$

$1$

$1$

$1)$

$b=np.$array$([0\mathrm{.}1,0\mathrm{.}3])$

$y=np*array([0,0\mathrm{.}4,2])$

It must return:

LAD of $2\mathrm{.}0$

Gradient of $\left[\begin{array}{ccc}-1& ?& -1.\end{array}\right]$