python-pytorchHow can I use Python PyTorch without CUDA?
You can use Python PyTorch without CUDA by setting the device
to cpu
when creating the Tensors. For example:
import torch
x = torch.rand(5, 3)
print(x.device)
device = torch.device("cpu")
x = torch.rand(5, 3, device=device)
print(x.device)
Output example
cpu
cpu
import torch
imports the PyTorch library.x = torch.rand(5, 3)
creates a tensor of size (5, 3) on the default device, which is usually the GPU.print(x.device)
prints the device on which the tensor is located.device = torch.device("cpu")
creates a CPU device object.x = torch.rand(5, 3, device=device)
creates a tensor of size (5, 3) on the CPU device.print(x.device)
prints the device on which the tensor is located.
Helpful links
More of Python Pytorch
- How can I use Python and PyTorch to create a Zoom application?
- How can I use Yolov5 with PyTorch?
- How can I use Python and PyTorch to create an XOR gate?
- How can I compare Python PyTorch and Torch for software development?
- How do I plot a PyTorch tensor using Python?
- How can I use Python, PyTorch, and YOLOv5 to build an object detection model?
- How can I use Python and PyTorch together with Xorg?
- How can I enable CUDA support in Python PyTorch?
- How can I use PyTorch with Python 3.10?
- How can I use PyTorch with Python 3.11?
See more codes...