def train$($self$,$ X$\_$train: np$.$ndarray, y$\_$train: np$.$ndarray, weights: np$.$ndarray$)->$ np$.$ndarray:

$"""$Train the classifier.

Use the linear regression update rule as introduced in the Lecture.

Parameters:

X$\_$train: a number array of shape $($N$,$ D$)$ containing training data;

N examples with D dimensions

y$\_$train: a numpy array of shape $($N$,)$ containing training labels

$"""$

N$,$ D $=$ X$\_$train.shape