Day 5 - Linear Regression Numpy Code and Python

Python

 In the course today we learned about the following concepts:

1. Lists 
2. Dictionaries
3. Tuples
4. Sets
5. Booleans
6. Dealing with Files in Python
7. Iterating in a file

Linear Regression Numpy Code

The code was completed and it was:

import math

beta = beta_zero

cost_diff = 100

rmse =-1

for i in range(10000):

    old_rmse = rmse

    y_hatnew = x_data.dot(beta)

        

    y_diff =y_true.reshape(len(x_inputs),1) - y_hatnew

        

    rmse = math.sqrt(y_diff.T.dot(y_diff)/x_data.shape[0])

    print(i,":",rmse)

        

    if abs(rmse-old_rmse) < 0.000000000001:

        break

    

    derivative = 2*y_diff.T.dot(x_data)/x_data.shape[0]

    beta = beta+step*derivative.T

print(beta) 

The next task given to us was to implement the sklearn function of Linear Regression and to compare the results of our version and the sklearn function.