I have 20.04 LTS and was running yolov5 model as root ( to rule out permissins issues ) and the model starts fine, then the first epoch of training starts, nothing happens, then I see the word "killed" and it all stops.

linux is still running and the terminal is OK.

How would I find out why linux would kill a process?

1 Answer 1


I diagnosed the issue - out of memory ( 8GB memory ).

I used :

dmesg -T| grep -E -i -B100 'killed process'

And an out of memory message was displayed.

I determined my model training batch size was too large, so I dropped it from 32 to 4 , and then it started and ran OK.

Hope this helps someone.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .