Week 3 Programming - Exercise 6 Back Prop

Course Q&A Deep Learning Specialization Neural Networks and Deep Learning
1

I have a problem with exercise 6 of week 3 assignment
All the tests prior to this one passed, and I believe I have the right equations to calculate the derivatives but I’m getting an error that the value of dW1 is wrong.
I retrieve W1 and W2 using parameters ["…"] and similar for A1 and A2 using cache.
I then do the calculations below but I’m getting the error wrong value of dW1

[Moderator edit: solution code removed.]

Please advise

2

Hi, @Mohamed_Othman. Please note that you are not allowed to post your solution code on Discourse or any other public forum. It is a violation of the course honor code. You should instead post a snapshot of the “traceback”, i.e. the error log printed to the screen which indicates how you would trace back (yup) through the code and the function calls to the likely source of your error. Learning to do this is an integral part of doing ML/AI.

3

Apologies for that
Kindly find below the traceback

AssertionError Traceback (most recent call last)
in
7 print ("db2 = "+ str(grads[“db2”]))
8
----> 9 backward_propagation_test(backward_propagation)

~/work/release/W3A1/public_tests.py in backward_propagation_test(target)
187 assert output[“db2”].shape == expected_output[“db2”].shape, f"Wrong shape for db2."
188
→ 189 assert np.allclose(output[“dW1”], expected_output[“dW1”]), “Wrong values for dW1”
190 assert np.allclose(output[“db1”], expected_output[“db1”]), “Wrong values for db1”
191 assert np.allclose(output[“dW2”], expected_output[“dW2”]), “Wrong values for dW2”

AssertionError: Wrong values for dW1

4

Thanks, @Mohamed_Othman. Tracebacks are not created equally! This one is on the disappointing side because there is not a single call to another function in backward_propagation(...). Your code threw an AssertionError from within the backward_propagation_test(...) function which was imported in the first cell via from public_tests import *.

You can view all of the functions in the public_test.py model in Jupyter with: “File → Open …” and then click on public_tests.py. If you inspect this code, you will see that your dW's and db's passed the “type” and “shape” tests. The dW1 array is producing the wrong values. Good new and bad news. Your dimensions are correct, but the values are not. Please recheck your expressions.