How to use PyTorch device and cuda.device
This article mainly explains "how to use PyTorch device and cuda.device". The content in the article is simple and clear, and it is easy to learn and understand. Please follow the editor's train of thought to study and learn how to use PyTorch device and cuda.device.
1 View current device input: import torchprint ("Default Device: {}" .format (torch.Tensor ([4,5,6]) .device))
Output:
Default Device: cpu
2 cpu devices can be specified using "cpu:0"
Input situation
Device= torch.Tensor ([1,2,3], device= "cpu:0") .deviceprint ("Device Type: {}" .format (device))
Output situation
Device Type: cpu
3 gpu devices can be specified using "cuda:0"
Input situation
Gpu = torch.device ("cuda:0") print ("GPU Device: [{}: {}]" .format (gpu.type, gpu.index))
Output situation
GPU Device: [cuda:0]
4 query the number of CPU and GPU devices
Input situation
Print ("Total GPU Count: {}" .format (torch.cuda.device_count () print ("Total CPU Count: {}" .format (torch.cuda.os.cpu_count ()
Output situation
Total GPU Count: 1
Total CPU Count: 8
5 convert from CPU device to GPU device 5.1 torch.Tensor method uses CPU device by default
Input situation
Data = torch.Tensor ([[1,4,7], [3,6,9], [2,5,8]) print (data.shape)
Output situation
Torch.Size ([3,3])
5.2 use the to method to convert the Tensor of cpu to the GPU device
Input situation:
Data_gpu = data.to (torch.device ("cuda:0")) print (data_gpu.device)
Output:
Cuda:0
5.3 use the .cuda method to convert the Tensor of cpu to the GPU device
Input situation:
Data_gpu2 = data.cuda (torch.device ("cuda:0")) # if there is only one piece of gpu, write it directly like this: data_gpu2 = data.cuda () print (data_gpu2.device)
Output:
Cuda:0
Thank you for your reading, the above is the content of "how to use PyTorch device and cuda.device". After the study of this article, I believe you have a deeper understanding of how to use PyTorch device and cuda.device, and the specific use needs to be verified in practice. Here is, the editor will push for you more related knowledge points of the article, welcome to follow!
