Closed SHIDi233 closed 1 week ago
This version is not in the web () list. Go to the history list -> https://download.pytorch.org/whl/torch_stable.html Find out your CUDA version. For example the 'CUDA 11.3' is equals to 'cu113/'
Input in bash as follows:
# torch
pip install https://download.pytorch.org/whl/cu113/torch-1.12.1%2Bcu113-cp39-cp39-win_amd64.whl
# torchaudio
pip install https://download.pytorch.org/whl/cu113/torchaudio-0.12.1%2Bcu113-cp39-cp39-win_amd64.whl
# torchvision
pip install https://download.pytorch.org/whl/cu113/torchvision-0.13.1%2Bcu113-cp39-cp39-win_amd64.whl
If you have completed all commands, you could test it by using following python sentences:
import torch
torch.cuda.is_available()
torch.cuda.device_count()
If the output are True and a number>=1, illustrating we have installed successfully.
My CUDA version is 11.3.