Iou score: There was a problem grading your submission. Check stderr for more details

[fdavidsen8]

Advanced Computer Vision with TensorFlow

Link: Coursera | Online Courses & Credentials From Top Educators. Join for Free | Coursera

I was submitting my solution to the grader but it gave me the following:

I had no clue on what went wrong. What should I do?

Thanks in advance

[gent.spah]

Which course and specialization is this?

[fdavidsen8]

It is Advanced Computer Vision with TensorFlow, TensorFlow: Advanced Techniques Specialization, week 3 programming assignment.

[gent.spah]

Did you sort the problem out?

[fdavidsen8]

Yes, I have reached the part of calculating the final grade on the Colab.

[fdavidsen8]

So what should I do? I mean I have completed all the problem in the Colab, but my problem regarding the grader still persist.

[gent.spah]

Whats the grader saying, is there some other verbal error than just that!

[fdavidsen8]

Nothing, I would have solved the problem if the grader told me what cause the error.

[gent.spah]

I am not sure either but you should check these 2 posts:

CC3_W1_Assignment birds.h5 issue

For anyone who runs into this problem: Since staff is on break until January 2nd, and this also might take a little while to fix, I think this hacky work-around should be enough to make the autograder happy: First, in order to use the GPU in Colab, do NOT run this cell near the beginning of the assignment: # Install packages for compatibility with the autograder !pip install tensorflow==2.8.0 !pip install keras==2.8.0 At the beginning of section 5.2 Save your model for grading, add these c…

TF-AT: Using the Fallback Runtime to Avoid Slow Training in Colab

Google Colab updates its runtime periodically to use newer versions of packages like Tensorflow. This can sometimes affect the use of GPUs for notebooks that require older versions. You will notice that when you reach the training part of the exercise. The lack of GPU will result in waiting several minutes per epoch. To get around it, please use the fallback runtime before running any of the code in your notebook. Here are the steps: Make sure that you are already connected to a GPU. You can c…

[fdavidsen8]

Thanks, the first post do the work

[alhimik]

I also have the same problem of grader error. My solution overall runs completely fine, achieves required loss and accuracy. I’ve also checked that auto-encoded images look completely fine.

I’ve tried just about everything I saw as suggested here - fallback runtime, strictly following the instructions for 2.8.0 tensorflow and keras installation - nothing works.

[Wendy]

@alhimik, the assignment was recently updated to include the steps to help you work around the issue with the version mismatch with the grader.

If you are using the new version of the assignment, it should be downloading the model using the file name, final_model.h5

Some things to try:

1. Make sure you are using the latest version of the assignment (the one that downloads filename final_model.h5

2. Make sure you have followed the instruction in the assignment to go to Runtime > Restart Session and wait for the kernel to reconnect after you have pip installed tensorflow and keras version 2.8.0, and before you load temp_model.h5 and save final_model.h5

3. Make sure when you submit the assignment, you choose the “Replace file” link and select the downloaded file named final_model.h5