CUDA: Difference between revisions
Line 27: | Line 27: | ||
;Notes | ;Notes | ||
* For machine learning, | * For machine learning, use Anaconda or Docker's CUDA since different versions of TensorFlow and PyTorch require different CUDA versions. | ||
You may need to add <code>LD_LIBRARY_PATH=/usr/local/cuda/lib64</code> to your environment variables.<br> | You may need to add <code>LD_LIBRARY_PATH=/usr/local/cuda/lib64</code> to your environment variables.<br> |
Revision as of 19:39, 17 July 2021
Installation
Linux
# Add NVIDIA package repositories
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin
sudo mv cuda-ubuntu2004.pin /etc/apt/preferences.d/cuda-repository-pin-600
sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub
sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/ /"
wget https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64/nvidia-machine-learning-repo-ubuntu2004_1.0.0-1_amd64.deb
sudo apt install ./nvidia-machine-learning-repo-ubuntu2004_1.0.0-1_amd64.deb
sudo apt update
# Install NVIDIA driver
sudo apt install nvidia-driver-465
# Reboot
# Check that GPUs are visible using nvidia-smi
sudo apt install cuda
# Install development and runtime libraries
sudo apt install libcudnn8 libcudnn8-dev
- Notes
- For machine learning, use Anaconda or Docker's CUDA since different versions of TensorFlow and PyTorch require different CUDA versions.
You may need to add LD_LIBRARY_PATH=/usr/local/cuda/lib64
to your environment variables.
You can also do this in PyCharm.
GCC Versions
nvcc
sometimes only supports older gcc/g++ versions.
To make it use those by default, create the following symlinks:
sudo ln -s /usr/bin/gcc-6 /usr/local/cuda/bin/gcc
sudo ln -s /usr/bin/g++-6 /usr/local/cuda/bin/g++
Alternatively, you can use -ccbin
and point to your gcc:
-ccbin /usr/local/cuda/bin/gcc