I wonder if multi-gpu testing is supported?
I've tried to set -b 8 -d 8 while evaluating:
python3 tools/track.py -f exps/example/dancetrack/yolox_x.py -c xxx/xxx.pth.tar -b 8 -d 8 --fp16 --fuse
but got an error:
2022-07-19 10:51:23 | ERROR | yolox.core.launch:219 - An error has been caught in function '_distributed_worker', process 'MainProcess' (26659), thread 'MainThread' (139709987980224)
...
...
RuntimeError: tensors.size() > 0 INTERNAL ASSERT FAILED at "../torch/csrc/distributed/c10d/reducer.cpp":1929, please report a bug to PyTorch.
Hi, thanks for your excellent work!
I wonder if multi-gpu testing is supported? I've tried to set
-b 8 -d 8
while evaluating:python3 tools/track.py -f exps/example/dancetrack/yolox_x.py -c xxx/xxx.pth.tar -b 8 -d 8 --fp16 --fuse
but got an error: