すっぴんのUbuntu 20.04 LTSに最短でCUDA入れるメモ（CUDA12.1編）

前回の一部改変

bleu48.hatenablog.com

すっぴんのUbuntu 20.04 LTSにCUDA12.1入れるメモ

（所謂自分用コピペ元）

まず以下を確認

https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/

sudo wget -O /etc/apt/preferences.d/cuda-repository-pin-600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin
sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/3bf863cc.pub
sudo add-apt-repository "deb http://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/ /"

sudo apt -y install cuda-12-1

export CUDA_PATH=/usr/local/cuda-12
echo 'export CUDA_PATH=/usr/local/cuda-12' >> ${HOME}/.bashrc
export LD_LIBRARY_PATH=/usr/local/cuda-12/lib64:${LD_LIBRARY_PATH}
echo 'export LD_LIBRARY_PATH=/usr/local/cuda-12/lib64:${LD_LIBRARY_PATH}' >> ${HOME}/.bashrc
export PATH=/usr/local/cuda-12/bin:${PATH}
echo 'export PATH=/usr/local/cuda-12/bin:${PATH}' >> ${HOME}/.bashrc

以下は機械学習系（TensorRTなど）

sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64/7fa2af80.pub
sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64/ /"

sudo apt-get install libnvinfer-dev libnvonnxparsers-dev libnvparsers-dev libnvinfer-plugin-dev

sudo apt-get install zip unzip

---

ということで今年の世界選手権ではCUDA12.1を使います。

前回より若干速度出ている気がします。