CW1 3rd assignment: Gradient checking - unexpected value error

Course Q&A Deep Learning Specialization Improving Deep Neural Networks: Hyperparameter tun
1

There is a mistake in the backward propagation! difference = 0.6623181774407311

AssertionError Traceback (most recent call last)
in
6 expected_values = [0.2850931567761623, 1.1890913024229996e-07]
7 assert not(type(difference) == np.ndarray), “You are not using np.linalg.norm for numerator or denominator”
----> 8 assert np.any(np.isclose(difference, expected_values)), “Wrong value. It is not one of the expected values”

AssertionError: Wrong value. It is not one of the expected values

I’m receiving the above error. I went through my code multiple times but couldn’t find what the actual problem is. Please let me know the work around if anyone has found the solution to this. Thanks!

3

I’m using np.linalg.norm values for the numerator and denominator

4

Please check DM from me for further instructions.

5

Thank you for sending me your code. Your implementation of the gradapprox[i] is incorrect. Check the formula again:

gradapprox[i] = \frac{J^{+}_i - J^{-}_i}{2 \varepsilon}

It’s J, not theta, right?

6

Thank you so much for the response and your time. This resolved my issue!