Image Segmentation with U-Net (March 2023)

This is for the Deep Learning Specialization, Course 4, Week 3, Assignment 2: Image Segmentation with U-Net

I’m passing all tests, but the grader is giving me 0/100. I’m following the special instructions about resetting and clearing the kernel, and saving.

The grader says: “Cell #3. Can’t compile the student’s code. Error: IndexError(‘list index out of range’)”

Also, I’m unable to train the model (though this is not graded). I’ve tried many times and the kernel always crashes after a few seconds.

Hi @Matthew_Cochran,

This is Cell # 3. Please follow the instructions provided for it in the markdown.

After you are done exploring, revert back to N=2 . Otherwise the autograder will throw a list index out of range error.

Cheers,
Mubsi

Thanks! This fixed it. So obvious in hindsight.

That will definitely fix the grader error. But did that fix also cure your problem with the kernel dying when you run the training? If not, one thing to check is whether you’ve added any print statements that cause the output of the training to be more voluminous that it would be by default. That can cause issues if the memory image of the notebook gets too large.