TF-AF Course 3 Week 3 Assignment Error

[Ejaz_Habib_Dar]

After submitting my model.h5 file I get an iou score of 0 out of 100 and get this error
“There was a problem grading your submission. Check stderr for more details.”

[gent.spah]

What error if any other does it give you? Also check similar posts in the forum, you might have the same issues as other posters!

[Ejaz_Habib_Dar]

Hello, thank you for replying and I haven’t seen anyone posting the same issue. If you have seen a similar issue then kindly please share the link.

[Yan_Du]

I got the same error message after submission. Please let me know if there is a solution. Thanks!

[Ejaz_Habib_Dar]

Okay bro. you also let me know in case you found a solution.

[Deepti_Prasad]

Send your notebook via personal DM. Did you make the changes I told Ejaz

[Deepti_Prasad]

Send your notebook via personal DM. Click on my name and then message.

Regards
DP

[Deepti_Prasad]

Hello @Ejaz_Habib_Dar Ejaz and @Yan_Du ,

please check the comment below, if you have the same error or require correction related to the post comment below, make changes and let me know if you see any improvement.

C3W3 Assignment training time

Hello @memoros77, corrections required. in def conv_block your data_format code should be the recall function of image_ordering and not image_ordering. in def FCN8 In your Block codes as instructions given We recommend keeping the pool size and stride parameters constant at 2. Please use unit also 2 and not 3 or all the conv_block in the same def FCN8 Upsample o above and crop any extra pixels introduced kindly remove the below codes use_bias = False, data_format=IMAGE_ORDERIN…

Regards
DP

[anoopebey]

Hi @Ejaz_Habib_Dar and @Yan_Du , I had the same error when using a different version of TF (the default version in the colab notebook is TF2.15 which is not compatible with the autograder currently) than the recommended TF 2.8 version. Please check this answer in this post and see if it works for you.

C3W3 Assignment training time

For anyone who is running into issues with slow training times for this assignment, Colab was recently changed to no longer support GPU with Tensorflow 2.8 - and since this assignment specifically loads TF 2.8 for compatibility with the autograder. No GPU makes the training MUCH slower. The staff will fix this issue when they get back from break in early January, but in the meantime: Anyone who is working on a local machine environment can use the elegant solution @anoopebey came up with (see…

[Ejaz_Habib_Dar]

thanks @Deepti_Prasad and @anoopebey my issue has been resolved and I passed the Assignment.

[Deepti_Prasad]

That’s great you debugged yourself.

Keep learning!!
Regards
DP