The formula for normalizing the input training data to mu=0 and variance=1 (slide 28) is given as xnorm = (x - mu) / sigma^2, but this is incorrect. The 0-mean input should be divided by the sqrt(sigma^2). In other words the correct formula for normalized inputs should be: X_{norm} = (X - \mu) / …

Course 2, Week 1 input normalization formula in slide 28 incorrect

rmwkwok March 15, 2023, 3:59am 4

Nice catch! If we check out the video version of the slide, we will see that the formula has been corrected there:

Cheers,
Raymond

Topic		Replies	Views
Clarification about the formula for sigma Improving Deep Neural Networks: Hyperparameter tun	5	550	June 6, 2022
C2W3 quiz - understand answer to Question 8 Improving Deep Neural Networks: Hyperparameter tun	8	630	June 1, 2022
Normalization Formulas Improving Deep Neural Networks: Hyperparameter tun	3	519	February 6, 2023
Week 2 programming assignment - exercise6 (normalize-rows) Improving Deep Neural Networks: Hyperparameter tun	1	495	December 15, 2022
Course 2 : Week 1 : Ex-4 Improving Deep Neural Networks: Hyperparameter tun	2	517	July 17, 2022