Anaconda installation
PyCharm installation
PyTorch installation
The three folders of cuDNN are copied to the CUDA directory. Check whether the installation is successful:
# 进入cuda的安装目录下bin目录,执行:
nvcc -V
# 进入cuda的安装目录下extras/demo_suite目录,执行:
banwidthTest.exe
deviceQuery.exe
# 查看gpu运行情况:
nvidia-smi
Check whether the installation is successful:
import torch
print("Hello PyTorch {}".format(torch.__version__)) # 查看torch版本
print(torch.cuda.is_available()) # 查看cuda是否可用