Unable to open lab of C2 W3

Hi I am doing Build, Train, and Deploy ML Pipelines using BERT. Currently I am at week 3 but can not open the assignment since it is always saying “the account is in cleanup” after the kernel suddenly died.

My coursera email: coldmist.turjo@gmail.com

With best regards
Mirza

@esanina can you help regarding this one.

@moderators I think it needs some help.

The lab launched but now the Jupyter server is not starting. I hope these issues solves soon.

Jupyter server is not running. Every time I want to start sagemaker studio its having the same issue even if I restart.

Hi @Mirza_Mohtashim_Alam !

I’ll report this to the staff team and try to get back to you asap.

1 Like

Thank you very much.

@Mirza_Mohtashim_Alam thank you for the messages. I have refreshed the account for you, so you can try to access again. However, I am not sure if you still had problems with the lab - I could see a full 100 scores grade for your account in C2W3.

Please let me know if you’ll have any other issues.

1 Like

Hi Esanina, It has been solved and I could fully complete the lab. Thank you very much for taking care of this.

With best regards,
Mirza

2 Likes

Hi I am having the same problem in C2 W3 Build, Train, and Deploy ML Pipelines using BERT - for the last 5 days I have been trying to access this lab without success. The lab either hangs or says “Account still in cleanup”. Any help would be much appreciated.

@James_Baird: could you manage to solve it after all?

@esanina it seems that when the lab is not completed in one go (within the given time of 2 hours), the kernel gets shut down and one can’t reenter the lab (by clicking on Start Lab) for a while and this notification pops up accordingly: account is still in cleanup!

Beside, there seems to be dependency issues in the very begging of the notebooks (installing required packages) for C2W2 and C2W3 after AWS env has been updated as in executing the following cell:

!pip install --disable-pip-version-check -q sagemaker==2.35.0 # c2w2 & c2w3
!conda install -q -y pytorch==1.6.0 -c pytorch # c2w2 
!pip install --disable-pip-version-check -q transformers==3.5.1 # c2w2 

causes errors and one can’t finish the assignment!

Please take a look at the issues people are currently posting in the course community page to provide us some official updates how to fix the issues!

Cheers,

@mrgransky thank you for the message. Please add the following code at the end of the pip installations:
!pip install -q protobuf==3.20.*
or download the updated notebook from the S3 bucket. That should help.