Cuda10.0 configuration pytorch1.7.0+monai0.9.0

1. pytorch install

No corresponding cuda10.0 The order of , reinstall 10.2 It's too much trouble , Try to use 10.2 Of cudatoolkit, The results are feasible .

# CUDA 10.2  （10.0 It works , But if the command says cudatoolkit=10.0,pytorch yes cpu Version of , It's hard to change ）
conda install pytorch==1.7.0 torchvision==0.8.0 torchaudio==0.7.0 cudatoolkit=10.2 -c pytorch
# test
python
import torch
x = torch.rand(5, 3)
print(x)
The output should be something similar to:
tensor([[0.3380, 0.3845, 0.3217],
        [0.8337, 0.9050, 0.2650],
        [0.2979, 0.7141, 0.9069],
        [0.1449, 0.1132, 0.1375],
        [0.4675, 0.3947, 0.1426]])
Additionally, to check if your GPU driver and CUDA is enabled and accessible by PyTorch, run the following commands to return whether or not the CUDA driver is enabled:
import torch
torch.cuda.is_available()

2. monai install
conda install numpy=1.21 (monai rely on numpy>=1.17,pytorch>=1.7.0)
pip install monai
# test
python
import monai
monai.config.print_config() 

The test passed