C1_W2_Linear_Regression , Compute_Gradient

Course Q&A Machine Learning Specialization Supervised ML: Regression and Classification
1

{moderator edit: image of code removed}

pic2
pic21426×885 61 KB

I am not able to understand what the error actually is & any explanation will be highly appreciated , to be mentioned i’am new to python programming.

2

Please do not post your code on the forum. That’s not allowed by the Code of Conduct.
If a mentor needs to see your code, we’ll contact you with instructions.

Hint: Part of your last two lines of code are incorrect.

image

1 Like
3

I am having a similar error computing the gradient and I’m hoping for further clarification.

I am receiving an assertion error, which as I understand relates to an operator/command not a value. I’ve pasted the error returned below:

AssertionError Traceback (most recent call last)
in
6 print(‘Gradient at initial w, b (zeros):’, tmp_dj_dw, tmp_dj_db)
7
----> 8 compute_gradient_test(compute_gradient)

~/work/public_tests.py in compute_gradient_test(target)
60 dj_dw, dj_db = target(x, y, initial_w, initial_b)
61 #assert dj_dw.shape == initial_w.shape, f"Wrong shape for dj_dw. {dj_dw} != {initial_w.shape}"
—> 62 assert dj_db == -2, f"Case 2: dj_db is wrong: {dj_db} != -2"
63 assert np.allclose(dj_dw, -10.0), f"Case 1: dj_dw is wrong: {dj_dw} != -10.0"
64

AssertionError: Case 2: dj_db is wrong: -12.0 != -2

Could I have messed up the array? Its shape is (4,1) and I believe it should only be (4, )

Thanks in advance.

4

The error means that your code in compute_gradient() is computing the wrong value for dj_db.

Your code returned -12.0, the expected value was -2.

5

The assert with the arrow and indentation (in this case it’s marked as line 62) is the one that was triggered:

image
image818×274 15.6 KB

An error about the size or shape of dj_dw would have been triggered by line 61.

6

Hi @Caroline3,

The error is telling you that your value for dj_db is not what was expected - that your compute_gradient function is returning -12 instead of the expected value of -2.

Go back and check your implementation for compute_gradient and see if you can spot what’s wrong and why it is returning -12. You might want to add some print statements temporarily to see what values it is calculating vs what you would expect.

Check the description just above the compute_gradient carefully to understand what calculations your code should make. If you’re still stuck, try clicking on the green “Click for hints” immediately after the compute_gradient cell. The first level will give you general hints, as well as some blue links you can click into to get more and more detailed hints.

7

Thank you for your detailed explanation of the error, I was able to find my mistake (a typo in my code) and fix it. While I feel confident in my calculus skills, I am new to Python and learning to interpret the error messages. Thanks!