Week 3 assignment: bug in the test

[EkaterinaSerikova]

Hello,

I would like to raise the two issued with Week 3 task:

1. In Section 4.3The Loop, task 4 - please pay attention that A2 is redefined in the test, and changes its dimensions. It becomes 1x3, which makes it impossible to use in backward propagation. In order to use it, it should be redefined again so that it has correct dimensions.

2. I passed all tests, but the grade is 75%, with the feedback “Tests failed on 2 cell(s)! These tests could be hidden. Please check your submission.”. With such feedback, it takes really lots of time to find a bug, especially when accuracy of the model matches the answer in the end.

If you could fix it or come back with some feedback regarding these two issues, this will be appreciated.

[Mubsi]

Hey @EkaterinaSerikova , thank you for this feedback. Fixing these are on our todo list.

[kenjiball]

@EkaterinaSerikova

A2

Has this been fixed? I have just been spending hours on the same task and getting test failed.

In the backward propagation, I get test failed but I have gone through the code so many times. I cant see the error.

(Solution code removed by staff)

[Mubsi]

Hi @kenjiball, yes, this has been fixed. Are you sure you are using the updated assignment version ? If not, follow the instructions on how to get it.

If you are still having issues even after this, drop me message with your lab ID.

You can find your lab ID in the url of the page when yoru assignment is open. The url will look something like this:
https://tcbcnayq.labs.coursera.org/tree/W4A1

Here, the first part will be your ID. In the example above, it is tcbcnayq

Hope it works,
Mubsi

[kenjiball]

Hey Mubsi,

I just tried updating the assignment but it didnt help. Maybe its something I am doing wrong. If you can take a look that would be great!

Here is the Url:
https://www.coursera.org/learn/neural-networks-deep-learning/programming/e6FsA/planar-data-classification-with-one-hidden-layer/lab

[Mubsi]

Hi @kenjiball, this is not the link I’m looking for.

Anyways, if you open your assignment, on the top right there will be a Help menu. Click on it, a side panel will appear, at the bottom of which it’ll show you your lab ID. Share that.

[kenjiball]

this is my lab code

mcxwwbzf

Thanks!

[kenjiball]

@Mubsi

Can someone please help me with this?

[Mubsi]

Hi @kenjiball

I have made changes to your notebook named Planar_data_classification_with_one_hidden_layer.ipynb

In there, I have also mentioned what your mistake was.

Apologise for replying late.

Best,
Mubsi

[kenjiball]

Thanks @Mubsi !!! Appreciate it

[dave_merit]

Please Help!
Im Getting a similar problem in labid bvkqtish
(wasting hours! Tried delete and reload … still no joy…)
To me its seems
my copy of public_tests.py line 30

def initialize_parameters_test(target):
n_x, n_h, n_y = 3, 5, 2

Conflicts with
Line 9 testCases_v2.py…

def initialize_parameters_test_case():
n_x, n_h, n_y = 2, 4, 1
return n_x, n_h, n_y

Even though is says " 1 Tests passed, 2 Tests failed"
My Output:

W1 = [[-0.00416758 -0.00056267]
[-0.02136196 0.01640271]
[-0.01793436 -0.00841747]
[ 0.00502881 -0.01245288]]
b1 = [[0.]
[0.]
[0.]
[0.]]
W2 = [[-0.01057952 -0.00909008 0.00551454 0.02292208]]
b2 = [[0.]]

appears identical to Expected Output:

W1 = [[-0.00416758 -0.00056267]
[-0.02136196 0.01640271]
[-0.01793436 -0.00841747]
[ 0.00502881 -0.01245288]]
b1 = [[0.]
[0.]
[0.]
[0.]]
W2 = [[-0.01057952 -0.00909008 0.00551454 0.02292208]]
b2 = [[0.]]

[Mubsi]

Hi @dave_merit, if you are still stuck here, I’d be happy to help. Kindly share your lab ID with me, thanks!

[dave_merit]

Lab ID is still bvkqtish

[fabovino]

HI Mubsi,
I experimented the same problem

mi ID fzwiabaa

Thanks

[sdidde]

Hello Mubsi,
I am seeing a similar response to dave_merit.
My code outputs the following result:

The expected result:

I updated the code to the latest version, rebooted the server, but no success.
Could you please help?
Here’s my ID: dtypwdeg
Thanks,
Stephen

[Mubsi]

Hi @sdidde,

You are passing in hardcoded values, for example, b1 = np.zeros((4,1)), here 4 and 1 are hardcoded values. You need to be passing in values dynamically.

Hint: read about the function’s parameters.

[sdidde]

Hello Mubsi,
Thanks for looking into my code.

[eeshita_pande]

Hello @Mubsi - getting the following output with the assignment grading tool:

All tests are passed in the notebook and lab id is tkzogije

Thanks in advance for your help.

[Mubsi]

Hi @eeshita_pande,

Do you still need help with this ?

Best,
Mubsi

[eeshita_pande]

Hi @Mubsi - yes, please. Thanks