Hello! I hope you are doing well.
This is Exercise 4 - gradient_check_n of week 1 of DLS course 2. I don’t know where I am making mistakes. Kindly guide me.
Saif.
PS: I will share my code here for a time being, if necessary.
Hello! I hope you are doing well.
This is Exercise 4 - gradient_check_n of week 1 of DLS course 2. I don’t know where I am making mistakes. Kindly guide me.
Saif.
PS: I will share my code here for a time being, if necessary.
Please click my name and message your notebook as an attachment.
Okay. I am sending it.
Saif.
Calculations of db1
and dW2
are incorrect in backward_propagation_n
.
Inside gradient_check_n
, calculation of theta_minus[i]
is incorrect. The name of the variable hints whether you should add or subtract epsilon
Oh, such a silly mistake. I am an idiot. Thanks a million, Balaji, for correcting me.
Regarding db1 and dW2, those incorrect equations are the part of the assignment. So, we can see that the Gradient check is high.
Once again, thanks a lot…
Saif.