单机多卡训练
问题1
Traceback (most recent call last):
File "train_tasks.py", line 493, in <module>
main()
File "train_tasks.py", line 237, in main
torch.distributed.init_process_group(backend="nccl")
File "/root/anaconda3/envs/vilbert/lib/python3.6/site-packages/torch/distributed/distributed_c10d.py", line 397,