下面两种python并行训练有什么不同啊？

我有2张显卡，怎么并行训练呢？？挖坑待填。。。。

# 来源：https://github.com/Tramac/awesome-semantic-segmentation-pytorch.git
import torch.nn as nn
local_rank = 0
model = nn.parallel.DistributedDataParallel(model, device_ids=[local_rank],  output_device=local_rank)

# 来源：RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching
from torch import nn
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = nn.DataParallel(model).to(device=device)