Hello everybody, thanks to Coursera for this opportunity
for assignment week 2 / in convolutional block :
I’ve got this error
what is the problem?
Hello everybody, thanks to Coursera for this opportunity
From the instructions,
To implement BatchNorm: BatchNormalization
BatchNormalization(axis = 3)(X, training = training). If training is set to False, its weights are not updated with the new examples. I.e when the model is used in prediction mode.
2 Likes
hello, thank you but I know the fact you just mentioned it, but I couldn’t solve the error.
is there any way to see my code?
Give me the lines where you specify training parameter to begin with 
1 Like
The other popular error on this section is to use the wrong input values for the shortcut section. They don’t really describe this very explicitly in the instructions for this section, but it is clearly shown in the diagrams.
2 Likes
@paulinpaloalto is right as always 
You have a mistake in the shortcut code. Check the diagrams and make sure the flow is correct Also, please remove your solution in the previous post when you are done.
1 Like
would you please guide me which part of shortcut code is not correct ?
Check against the diagram. Observe what is passed as input in each step.
2 Likes
thank you. sir
I found the answer.
1 Like
Thank you Paulin for answering my question.