Problem 3 Many ML models are parametrized. As we saw in the case study on house price prediction, the parameters of a multivariate linear regression model consist of its coefficients (also called weights) and intercept (also called bias). For a model with n variables or input features, we use here the notation w = [w, ..., wn] to denote the coefficients (weights) and use b to denote the intercept (bias) of the model. We call w the coefficient vector or the weight vector. On input a feature vector x = [x], ... , xn], the model would predict its value as linear_regression w.b(x) = Wixi + b In this exercise, you will implement this function in two ways. 3a) ) Write a function linear regression_sequential(w, b, x) that computes the above linear regression function using an explicit loop. Here the input w is a one-dimensional NumPy array representing the coefficient vector and b is a number representing the intercept of the linear model, and x is a one-dimensional NumPy array representing the input feature vector. In [ ]: def linear regression_sequential(w, b, x): " Implements the prediction function of a linear regression model using an explicit loop Args : w (ndarray): A one-dimensional Numpy array representing the coefficient vector of a linear model b (float): A number representing the intercept of a linear model x (ndarray): A one-dimensional Numpy array representing the input feature vector to the model Returns: float: A number representing the predicted output value of the input x assert w.ndim == 1 and x.ndim == 1 and w.shape == x.shape # Your code here