DLS C4 wk1pa1 - "All test passed!" but 75/100 ... Common fixes do not help

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

Hello !

I could finish the 1st programming assignement of week 1 of course 4 ( Convolutional Model, Step by Step), with “All test passed!” for every mandatory cells, in the jupyter notebook.
… but, i only get 75% score after submission. Frustrating :frowning:

I’ve got the hereunder info from the grader. It’s unclear. Is there a way to know which is the problematic cell ?

It seems like a common syndrome, but i couldn’t solve it using suggestions in similar topics : ensure using np.float in “conv_single_step,” and ensuring to compute the corners using the stride in pool_forward". None of the works for me.

Can someone help ? Thanks :slight_smile:

Regards.

[ValidateApp | INFO] Validating ‘/home/jovyan/work/submitted/courseraLearner/W1A1/Convolution_model_Step_by_Step_v1.ipynb’
[ValidateApp | INFO] Executing notebook with kernel: python3
Tests failed on 1 cell(s)! These tests could be hidden. Please check your submission.

2

okay. I could figure it out. probleme solved

Stride should be used for computing the corners in both pool_forward and conv_forward. I was using it only in the pool_forward.

Nevertheless, having the grader telling which cell doesn’t pass would be a time saver. :slight_smile:

Best regards

Denis.

3

Yes, it would be great if the grader could tell you which function was at fault, but apparently that is not so easy to fix.

What should be easy to fix and is in progress is making the test cases in the notebook better at failing when they should. But that will probably take another few weeks to get resolved.