Image Segmentation with U-Net Assignment

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

I am not sure even though for GRADED FUNCTION: unet_model I see All tests are passed.
I get low score and only feedback I receive is: Code Cell UNQ_C3: Function ‘unet_model’ is incorrect. Check implementation.

Code Cell UNQ_C1: Function ‘conv_block’ is correct.
Code Cell UNQ_C2: Function ‘upsampling_block’ is correct.
Code Cell UNQ_C3: Function ‘unet_model’ is incorrect. Check implementation.

2

It’s hard to write tests in the notebook that catch all possible errors. The most common way to pass the tests in the notebook, but fail the grader, is to “hard-code” something so that it happens to match the input data in the notebook.

1 Like
3

Hey @Maulik1,
Welcome to the community. In view of what Paul Sir has said, have you hard-coded the filters in the conv_block and upsampling_block? Do check it out, and if yes, define them in terms of n_filters. I hope this helps.

Regards,
Elemento

1 Like
4

Thank you, Paul and Vishesh.

Two plausible reasons could be I was calling conv_block with n_filters = 32 (which I changed to n_filters) And had added two additional lines with comments while debugging - which shouldn’t alter output any way but I removed that.
Any how, grader doesn’t complain any more.

Thank you again for taking time and helping.