Fill in the blanks Kadd_to Logisticregressiondef fit(self, x, Y, epochs-100e, print_loss=True):This function implements the Gradient Descent AlgorithmArguments:x -- training data matrix: each column is a training example.The number of columns is equal to the number of training examplesY -- true "label" vector: shape (1, m)epochs --Return:params --dictionary containing weightslossesloss values of every 100 epochs--grads --dictionary containing dw and dw_elosses = []for i in range(epochs):# Get the number of training examplesm = x. shape[1]### START YOUR CODE HERE #### Calculate the hypothesis outputs A (* 2 lines of code)Z =A =# Calculate loss (* 1 line of code)loss =# calculate the gredients for w and w_edw =dw_e =# weight updatesself.W =self.w_e =### YOUR CODE ENDS ###

Answered: Kadd_to LogisticRegression def…

Database System Concepts

7th Edition

ISBN:9780078022159

Author:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Publisher:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Chapter1: Introduction

Section: Chapter Questions

Problem 1PE

See similar textbooks

Related questions

Question

Fill in the blanks

Kadd_to Logisticregression
def fit(self, x, Y, epochs-100e, print_loss=True):
This function implements the Gradient Descent Algorithm
Arguments:
x -- training data matrix: each column is a training example.
The number of columns is equal to the number of training examples
Y -- true "label" vector: shape (1, m)
epochs --
Return:
params --
dictionary containing weights
losses
loss values of every 100 epochs
--
grads --
dictionary containing dw and dw_e
losses = []
for i in range(epochs):
# Get the number of training examples
m = x. shape[1]
### START YOUR CODE HERE ###
# Calculate the hypothesis outputs A (* 2 lines of code)
Z =
A =
# Calculate loss (* 1 line of code)
loss =
# calculate the gredients for w and w_e
dw =
dw_e =
# weight updates
self.W =
self.w_e =
### YOUR CODE ENDS ###

Expert Solution

Step 1

%%add_to MultivariateNetwork
def fit(self, X, Y, epochs=1000, print_loss=True):
    
    losses = []

    print("W.T.shape = ", self.W.T.shape)
    print("X.shape = ", X.shape)
    print("Y.shape = ", Y.shape)
    print("W.shape = ", self.W.shape)

   
    
    
    for i in range(epochs):

      ### START YOUR CODE HERE ### 

      m = X.shape[1]

     
      
      Y_hat = np.dot(self.W.T, X)
      
      # Calculate loss (≈ 1 line of code)
      loss = (1/(2*m)) * np.dot((Y_hat - Y), (Y_hat - Y).T)
      
      # Calculate the gredients for W and w_0 (≈ 2 lines of code)
      # dw = np.dot(np.sum(Y_hat - Y), X)
      dw = np.sum((Y_hat - Y) * X)
      dw_0 = np.sum(Y_hat - Y)
      
      dw /= m
      dw_0 /= m

      # Weight updates (≈ 2 lines of code)
      self.W -= self.alpha * dw
      self.w_0 -= self.alpha * dw_0
      ### YOUR CODE ENDS ###