[Week 3] Image segmentation U-net assignment

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

Hi,
I passed all the internal tests in the notebook as all the test blocks outputted "All tests passed!’ but for some reason when submitting I’m getting 0/100 with the following grader output message “Error while grading submission. Please check your submission.”
Also the predictions looks funny so am I doing something wrong?

Could you guide me on how to fix this as it’s very tricky given I passed all the unit tests!

Screenshot 2021-04-24 161302
Screenshot 2021-04-24 161302579×502 43.5 KB

Update: When submitting the assignment later on, it passed! It must have been a network problem or so :man_shrugging:t3:. But I got 75% accuracy training this net for 3 epochs and the predictions seem way off for me! I’ll keep investigating anyway :grin:!

2

I am having the same problem. Despite passing all the tests, I get 0 points from submission.

Moreover, although I was submitting at different times, the problem was still not solved.
Before each submisson I reset the kernel and rerun the codes. I would be very happy if this issue is dealt with.

Thank you from now.

3

I opened a case with coursera’s support first then tried to resubmit on my own afterwards. I didn’t get their feedback though. See if that helps.

4

Hi @fady_michel and @hanseref,

I have reported this issue to people working on the backend of deeplearning.ai.

I myself had to submit three times to get the notebook to pass (so it did pass in the end).

The predictions look strange, but this is because the model should be trained much longer; that would take too long for this course.

I hope the issue will be resolved soon, or that you manage to pass with the submission after all.

Thanks for reporting!

1 Like
5

Same problem here. Here is the log I got from the grader. Hopefully it may shed some light.

[ValidateApp | INFO] Validating ‘/home/jovyan/work/submitted/courseraLearner/W3A2/Image_segmentation_Unet_v1.ipynb’
[ValidateApp | INFO] Executing notebook with kernel: python3
2021-04-25 23:25:27.048790: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘libcudart.so.10.1’; dlerror: libcudart.so.10.1: cannot open shared object file: No such file or directory
2021-04-25 23:25:27.048882: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2021-04-25 23:25:31.271714: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘libcuda.so.1’; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory
2021-04-25 23:25:31.271781: W tensorflow/stream_executor/cuda/cuda_driver.cc:312] failed call to cuInit: UNKNOWN ERROR (303)
2021-04-25 23:25:31.271814: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (somehost): /proc/driver/nvidia/version does not exist
2021-04-25 23:25:31.272161: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2021-04-25 23:25:31.281927: I tensorflow/core/platform/profile_utils/cpu_utils.cc:104] CPU Frequency: 2999995000 Hz
2021-04-25 23:25:31.283579: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7efc38623090 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2021-04-25 23:25:31.283622: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
[ValidateApp | ERROR] Timeout waiting for execute reply (30s).
[ValidateApp | ERROR] Interrupting kernel
2021-04-25 23:26:36.400987: W tensorflow/core/kernels/data/cache_dataset_ops.cc:798] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to dataset.cache().take(k).repeat(). You should use dataset.take(k).cache().repeat() instead.
2021-04-25 23:26:36.655239: W tensorflow/core/kernels/data/cache_dataset_ops.cc:798] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to dataset.cache().take(k).repeat(). You should use dataset.take(k).cache().repeat() instead.
Success! Your notebook passes all the tests.

6

Hey @vuqpham if you look at the very last line it says all tests passed. You should have gotten a full score. Can you confirm ?

As for the message itself, I suspect it is because code blocks have timeout on them, when it runs out it displays back as this log you are seeing, but also passing all the tests. The reason you are able to pass the tests despite this log is because these code blocks (against which you are getting this) are not required for grading.

7

Oh no, I got 0/100. Yes, I did notice the last line, but with too many error logs, I believe the grader never grades my notebook, and somehow (some kind of "if then else " bug :slight_smile: ) prints out that line.

8

Hey @vuqpham this seems to be a random occurrence. I have submitted the same code, first six times it was 100/100, then the next two 0/100 with “Error while grading submission. Please check your submission.” But the next time 100/100 again.

Unfortunately I don’t have a lot to offer at time this point, expect that keep trying by submitting over and over again. I’ll look into this on Monday. We apologise for the inconvenience.

P.S a learner did this and it worked for him. Maybe if could work for you as well

  • Restart Kernel and clear all output
  • run all necessary codes from begining to last graded function and its test
  • do not run remaining codes
  • submit
9

Thanks Mubsi :+1:. I will try and report back.

1 Like
10

I got 100/100 twice. Here is what I did:
1.

  • restart kernel and run all
  • submit
  • restart kernel and clear output (and not run any cell)
  • submit.

Both cases got 100/100, but the grader still showed a lot of errors like the first time.

Below is the log of the latest submission.

[ValidateApp | INFO] Validating ‘/home/jovyan/work/submitted/courseraLearner/W3A2/Image_segmentation_Unet_v1.ipynb’
[ValidateApp | INFO] Executing notebook with kernel: python3
2021-04-26 01:35:07.230607: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘libcudart.so.10.1’; dlerror: libcudart.so.10.1: cannot open shared object file: No such file or directory
2021-04-26 01:35:07.230645: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2021-04-26 01:35:09.076815: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library ‘libcuda.so.1’; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory
2021-04-26 01:35:09.076853: W tensorflow/stream_executor/cuda/cuda_driver.cc:312] failed call to cuInit: UNKNOWN ERROR (303)
2021-04-26 01:35:09.076878: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (somehost): /proc/driver/nvidia/version does not exist
2021-04-26 01:35:09.077169: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2021-04-26 01:35:09.082569: I tensorflow/core/platform/profile_utils/cpu_utils.cc:104] CPU Frequency: 2999995000 Hz
2021-04-26 01:35:09.083291: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7fd896e39d00 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2021-04-26 01:35:09.083322: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
[ValidateApp | ERROR] Timeout waiting for execute reply (30s).
[ValidateApp | ERROR] Interrupting kernel
2021-04-26 01:36:05.768188: W tensorflow/core/kernels/data/cache_dataset_ops.cc:798] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to dataset.cache().take(k).repeat(). You should use dataset.take(k).cache().repeat() instead.
2021-04-26 01:36:05.975719: W tensorflow/core/kernels/data/cache_dataset_ops.cc:798] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to dataset.cache().take(k).repeat(). You should use dataset.take(k).cache().repeat() instead.
Success! Your notebook passes all the tests.

11

I’ve done all your suggestions but in my case the issue persist and I’ve got 0/100 grade.
In my case at the end of file doesn’t appear any score or grade.

12

I am experiencing the same problem. I have submitted 10 times, with and without cleaning the cells, and nothing. Any help?

EDIT: I fixed it, my problem was that I changed the value of N in cell 3, and that was the problem.