出错背景：在我的训练过程中，因为任务特殊性，用的是多卡训练单卡测试策略。模型测试的时候，由于数据集太大且测试过程指标计算量大，因此测试时间较长。
报错信息：
File "/home/anys/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 940, in __init__self._reset(loader, first_iter=True)File "/home/anys/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 971, in _resetself._try_put_index()File "/home/anys/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 1205, in _try_put_indexindex = self._next_index()File "/home/anys/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 508, in _next_indexreturn next(self._sampler_iter)# may raise StopIterationFile "/home/anys/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/utils/data/sampler.py", line 227, in __iter__for idx in self.sampler:File "/home/anys/GRALF/AlignTransReID/TransReID/datasets/sampler_ddp.py", line 148, in __iter__seed = shared_random_seed()File "/home/anys/GRALF/AlignTransReID/TransReID/datasets/sampler_ddp.py", line 108, in shared_random_seedall_ints = all_gather(ints)File "/home/anys/GRALF/AlignTransReID/TransReID/datasets/sampler_ddp.py", line 77, in all_gathergroup = _get_global_gloo_group()File "/home/anys/GRALF/AlignTransReID/TransReID/datasets/sampler_ddp.py", line 18, in _get_global_gloo_groupreturn dist.new_group(backend="gloo")File "/home/anys/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/distributed/distributed_c10d.py", line 2503, in new_grouppg = _new_process_group_helper(group_world_size,File "/home/anys/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/distributed/distributed_c10d.py", line 588, in _new_process_group_helperpg = ProcessGroupGloo(RuntimeError: Socket Timeout从报错信息中可以看到是数据加载的时候，创建进程引起的超时，解决方法就是将“进程”的“存活”时间加大：
torch.distributed.new_group(backend="gloo",timeout=datetime.timedelta(days=1))当出现超时报错时，可以先检查所有创建子进程的地方，将timeout调大。