Can someone tell me how to get pass the complete script when experiencing breakdown from Kernel?
I was going through 2.2 - Fine-Tune the Model with the Preprocessed Dataset and whenever I try to run the trainer.train() code, Kernel will automatically restart and the previous steps are complete gone for no reasons. I have been trying different ways such as restarting the search engine or lab 2 pages, but nothing works, and the code is not graded either.
Please help and thanks
There seems to be an issue on the AWS end, I am not sure if it impacts you but might be the case, so hang on for a couple of hours.
I was getting same error trainer.train() Kernel Restart. Also notice that I can not set
8 vCPU+32GIB kernel dependencies only able to set 2vCPU+4GIB. Other selections do not work.
OK - I was able to set the vCPU8 on looking at this setting there the list had fast and the all instances - In the All instance could get to the vCPU8. I set this and then was able to get past the restart error for trainer.
Hi, I am having same kernel restart issue, even after 2 weeks. Please help.
Make sure your CPU and GiB are set to correct vCPU 8 and 32 GIB. It’s not the first category but the All instances has the version you want. So the drop down has multiple drop down for setting the CPU and GIB - pick the second all Instances and then pick vCPU 8 and 32 GIB within that All instances.
I have tried to select your suggestion, but I’m not allowed to.
I am also facing the same issue, Any workaround for this issue?
Changing the instance type to “ml.m5.2xlarge” worked for me.
Clic on corner right text “data science etc…” and change the instance type so you have 8vGPU