Simple and fast cuda11.7 configuration and torch2.0 installation
-
- 1. cuda environment configuration
- 1. cuda下载-https://developer.nvidia.com/cuda-toolkit-archive
- 2. Download cudnn-https://developer.nvidia.com/rdp/cudnn-archive corresponding to cuda
- 3. For the cuda installation process [no brain installation - all press yes] - NVIDIA GPU Computing Toolkit will appear in the program files file of the C drive
- 4. Copy the bin, include, and lib packages under the cudnn file corresponding to cuda to the bin, include, and lib packages under the corresponding cuda folder respectively
- 5. nvidia-smi test
- 2. Create an environment in anaconda
- 1. Create an environment
- 2. Enter the pytorch official website page - https://pytorch.org/get-started/locally/ copy the download command
1. cuda environment configuration
1. cuda下载-https://developer.nvidia.com/cuda-toolkit-archive
2. Download cudnn-https://developer.nvidia.com/rdp/cudnn-archive corresponding to cuda
3. For the cuda installation process [no brain installation - all press yes] - NVIDIA GPU Computing Toolkit will appear in the program files file of the C drive
4. Copy the bin, include, and lib packages under the cudnn file corresponding to cuda to the bin, include, and lib packages under the corresponding cuda folder respectively
For example, the bin package files in -cudnn are all copied to the bin in cuda
5. nvidia-smi test
2. Create an environment in anaconda
1. Create an environment
conda create -n AI_2 python=3.8
activate AI_2
2. Enter the pytorch official website page - https://pytorch.org/get-started/locally/ copy the download command
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117```