Week 1, Gradient Checking, Excercise 4, gradient_check_n

There is no bug to fix, and I rechecked my code but the outcome of the function is still too big. Any suggestion for the solution, please.

Eventhough it is too big, but i still pass the test 100/100. I wonder why

Nice work! It looks like your code is correct. Please check the assertion statement in that block of code more carefully. Your value is one of the two correct values. Your value is the one you get before you remove the intentional silly bugs that they put in the back propagation routine. If you remove them, then you’ll get the other (much smaller) answer. So what this demonstrates is that your gradient checking code successfully detected the fact that there was in fact a bug in the back propagation logic. :nerd_face:

Nevermind, I didn’t read the hints properly. Note to self- do not needlessly rush

1 Like

It’s great to hear that you found the solution and realized the “meta” lesson there as well! There’s a saying in English that I think captures your point: “Go slow to go fast.” Meaning that “saving time” by not reading the instructions carefully usually ends up not being a net savings of time. You save a couple of minutes and then waste an hour or two trying to figure out why things didn’t come out the way they should. :nerd_face: