The Jupyter notebook kernel runs in the browser and has memory limitations compared to when running on desktop / the cloud. Out of memory is the likely the cause of kernel crash, but to confirm it would be great with a reproducible example.

Hello @Anders Hafreager , 

Thanks for your reply. I am facing this issue often. Also, my observation is - it crashes when ‘ML models are used for training’. 

Do you require any type of logs / screen recording to check upon this ?

 

You can check the browser memory usage using Chrome profiler, but ideally if you can share a notebook I can check also

@Anders Hafreager will PM the notebook file which Ankit shared.

Hi @Ankit Kothawade,

Would you be able to give this a try in another notebook?

Hi @Ankit Kothawade,

I hope the issue has been resolved. I'm closing this topic for now. Please feel free to start a new post if you have any questions.