class Optimize:

def __init__(self, learning_rate=1e-4, reg=1e-3):

self.learning_rate = learning_rate

self.reg = reg

def update(self, model):

pass

def apply_regularization(self, model):

'''

Apply L2 penalty to the model. Update the gradient dictionary in the model

:param model: The model with gradients

:return: None, but the gradient dictionary of the model should be updated

'''

#############################################################################

# TODO: #

# 1) Apply L2 penalty to model weights based on the regularization #

# coefficient #

Weights are in a dictionary as you. an see example below

self.weights['W1'] = 0.001 * np.random.randn(inputsize, n_classes)

self.gradients['W1'] = np.zeros((inputsize, n_classes))

#############################################################################

Need to solve the function in python as instructions mention and having a hard time