Course4_week_4_exercise(1)_autograder_issue

Course Q&A Deep Learning Specialization Convolutional Neural Networks
,
1

The autograder keeps giving me 0 points out of the total even though my output matches the predicted output.

2

Hi @AshishNair

Getting outputs just like the predicted output does not guarantee passing the grader! Please make sure your code is handling edge cases and avoid hard-coding, this should help!

Feel free to ask if you need further assistance.

3

Hi
I cannot seem to find any errors as such and am pretty confident that it is supposed to work, I don’t seem to find the “All tests passed” being displayed as it did all the other assignments despite the logic being correct and matching with the corresponding outputs. Is there any chance that it could be something related to the autograder?

4

If the grader gives you 0 points, there is something wrong with your notebook. Either your code is wrong or there is something structurally wrong with your notebook. We need to debug this situation. The first step is to show us the actual grader output under “Show grader output”. Please copy/paste that or post a screen shot of it.

5

The tests in the notebook do not prove your code is perfect.
The grader uses different tests, and a different execution context.

6

image
image974×680 10.5 KB

7

Hi @AshishNair ,

The comment line ‘UNQ_C1’ has an assignment key which the grader failed to find it in your notebook. It could be that you have accidentally removed that line, causing the grader to fail.

You can get a fresh copy of the lab and redo the assignment:
First - rename the current copy for reference. No fresh copy will be loaded if there is a file with the same file name in the directory.
Second - click on ‘?’ button on the top right of the Coursera window and follow the instruction.

1 Like
8

Yes, as Kin says, you must have modified the comments in at least one of the function cells. Here’s what the first few lines of the first graded cell should look like:

# UNQ_C1(UNIQUE CELL IDENTIFIER, DO NOT EDIT)
# GRADED FUNCTION: triplet_loss

def triplet_loss(y_true, y_pred, alpha = 0.2):

If you removed those comment lines, then the grader will fail. Note that they even warned you not to mess with that line. :smile: It is generally a mistake to modify things outside the “YOUR CODE HERE” sections. It’s not against the rules to add code outside those areas, but you need to be very careful that you know what you’re doing when you do that.

If you need more details about how to get a fresh copy of the notebook, here’s a thread about that.