software dependence error about torchseg HOT 5 CLOSED

Hi, my software environments are as follows:

Ubuntu 16.04
PyTorch 1.0
Python 3.5.2
Cuda 9.0
GCC 5.4.0

According to my experimence, the warning you showed is normal. Then you should install the apex to use the multi-process-sync-bn.

from torchseg.

hi, @ycszen.
I got a similar issue, when I run "python train.py -d 0-2", it gives a segmentation fault.
Do you have any idea ?

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

                          !! WARNING !!

platform=sys.platform))
/home/daodao/anaconda3/lib/python3.6/site-packages/torch/utils/cpp_extension.py:166: UserWarning:

                           !! WARNING !!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Your compiler (c++) is not compatible with the compiler Pytorch was
built with for this platform, which is g++ on linux. Please
use g++ to to compile your extension. Alternatively, you may
compile PyTorch from source using c++, and then you can also use
c++ to compile your extension.

See https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md for help
with compiling PyTorch from source.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

                          !! WARNING !!

platform=sys.platform))
13 10:32:26 PyTorch Version 1.0.0, Furnace Version 0.1.1
13 10:32:26 using devices 0, 1, 2
[00:00<?,?it/s]Segmentation fault (core dumped)

from torchseg.

thanks very much @ycszen
I have successfully run the script.

from torchseg.

@daodaofr Did you solve this question? I meet the same one

from torchseg.

@Mr-doooog I was using the wrong labels (*_color.png), you need to use *_labelledIds.png.

from torchseg.