[arcface_torch]arcface pytorch training is working ok with 2 gpus , but failed at 4 gpus by my setting , details shown below

while i trained the arcface pytorch with 2 gpus , which used the command below , it worked fine :
`python -m torch.distributed.launch --nproc_per_node=2 --nnodes=1 --node_rank=0 --master_addr="127.0.0.1" --master_port=1234 train.py`

but failed by set the nproc_per_node=4 , my command is :
`python -m torch.distributed.launch --nproc_per_node=4 --nnodes=1 --node_rank=0 --master_addr="127.0.0.1" --master_port=1234 train.py`

the error is:
```
File "train.py", line 135, in
Traceback (most recent call last):
File "train.py", line 135, in
main(args_)
File "train.py", line 26, in main
rank=args.local_rank)
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/distributed_c10d.py", line 455, in init_process_group
barrier()
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/distributed_c10d.py", line 1960, in barrier
main(args_)
File "train.py", line 26, in main
rank=args.local_rank)
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/distributed_c10d.py", line 455, in init_process_group
barrier()
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/distributed_c10d.py", line 1960, in barrier
work = _default_pg.barrier()
RuntimeError: NCCL error in: /pytorch/torch/lib/c10d/ProcessGroupNCCL.cpp:784, unhandled system error, NCCL version 2.7.8
work = _default_pg.barrier()
RuntimeError: NCCL error in: /pytorch/torch/lib/c10d/ProcessGroupNCCL.cpp:784, unhandled system error, NCCL version 2.7.8
Traceback (most recent call last):
File "train.py", line 135, in
main(args_)
File "train.py", line 26, in main
rank=args.local_rank)
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/distributed_c10d.py", line 455, in init_process_group
barrier()
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/distributed_c10d.py", line 1960, in barrier
work = _default_pg.barrier()
RuntimeError: NCCL error in: /pytorch/torch/lib/c10d/ProcessGroupNCCL.cpp:784, unhandled system error, NCCL version 2.7.8
Traceback (most recent call last):
File "/opt/conda/lib/python3.6/runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
File "/opt/conda/lib/python3.6/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/launch.py", line 260, in
main()
File "/opt/conda/lib/python3.6/site-packages/torch/distributed/launch.py", line 256, in main
cmd=cmd)
subprocess.CalledProcessError: Command '['/opt/conda/bin/python', '-u', 'train.py', '--local_rank=3']' returned non-zero exit status 1.
```

@anxiangsir do you have any idea ? great thanks !