search

zhengye1995 / Tianchi-2019-Guangdong-Intelligent-identification-of-cloth-defects-rank5

天池2019广东工业智造创新大赛 布匹疵点检测 天池水也太深了 季军解决方案
403 stars 142 forks source link

"CalledProcessError" when training data on my workstation #6

Closed YuzhouPeng closed 4 years ago

YuzhouPeng commented 4 years ago

I use a new machine(CentOs7) and the program return CalledProcessError when running dist_train.sh: 捕获 this is train.sh file: 捕获1 and dist_train.sh: 捕获2 All dependencies install succeed. cudnn version 7.6.3

Could you please help me figure out the problem? Thank you!

YuzhouPeng commented 4 years ago

I use a single RTX 2080 on my machine

zhengye1995 commented 4 years ago

Hi @YuzhouPeng , Signals.SIGSEGV: 11 seems to be a common bug when you are using different libraries/environments for compiling and running the code. I find a similar issue in other projects such as nvvl, this might give you a hint to check the running/compiling libraries to fix the bug.

And I think you can open a issue in mmdetection , maybe you can find some way to fix it.

YuzhouPeng commented 4 years ago

Thank you!