Skip to content

Test the Cifar10 dataset on a single GPU and multiple GPUs to compare their performance differences

Notifications You must be signed in to change notification settings

YourPeer/pytorch_distributed_training

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

pytorch_distributed_training

测试步骤:

1、创建data文件夹
2、将代码文件里的CIFAR10函数中的download参数改为True

  trainset=CIFAR10(data_dir, train=True, transform=transform, target_transform=None, download=True)  
  testset = CIFAR10(data_dir, train=False,download=True, transform=transform)

测试结果:

1、在单个gpu的情况下,训练一个epoch的精度可以达到百分之50,但训练时间为249秒。

simple_gpu.png

2、在六个gpu的情况下,训练一个epoch的精度虽然只有百分之41,但训练时间仅为115秒。

multi_gpu.png

About

Test the Cifar10 dataset on a single GPU and multiple GPUs to compare their performance differences

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages