C2_W3_Assignment: Error calculation for model evaluation

Course Q&A Machine Learning Specialization Advanced Learning Algorithms
1

Hello!

Why in ‘C2_W3_Assignment: Practice Lab: Advice for Applying Machine Learning’ Mean Squared Error (part 2.2 Error calculation for model evaluation, linear regression. Exercise 1) is calculated by dividing the sum of squared errors by 2*m but not by m examples like in Scikit-learn metrics?

2

Hi @VeronikaS

Divide the sum of squared errors by 2m not by m isn’t make abig difference, But We divide by 2m because when we calculate the gradient descent(Back propagation) it make the calculation more easier becausethe derivative of the power(square) remove the division by 2 so that the calculation became more faster and easier

Best Regards,
Abdelrahman

1 Like
3

@AbdElRhaman_Fakhry thank you for help!