sudo apt update && sudo apt upgrade
sudo apt autoremove nvidia* --purge
ubuntu-drivers devices
You will install the NVIDIA driver whose version is tagged with recommended
sudo ubuntu-drivers autoinstall
My recommended version is 525, adapt to yours
sudo apt install nvidia-driver-525
reboot
after restart verify that the following command works
nvidia-smi
sudo apt update && sudo apt upgrade
sudo apt install nvidia-cuda-toolkit
nvcc --version
You can download cuDNN file here. You will need an Nvidia account. Select the cuDNN version for the appropriate CUDA version, which is the version that appears when you run:
nvcc --version
sudo apt install ./<filename.deb>
sudo cp /var/cudnn-<something>.gpg /usr/share/keyrings/
My cuDNN version is 8, adapt the following to your version:
sudo apt update
sudo apt install libcudnn8
sudo apt install libcudnn8-dev
sudo apt install libcudnn8-samples
sudo apt-get install python3-pip
sudo pip3 install virtualenv
virtualenv -p py3.10 venv
source venv/bin/activate
pip3 install torch torchvision torchaudio
import torch
print(torch.cuda.is_available()) # should be True
t = torch.rand(10, 10).cuda()
print(t.device) # should be CUDA
Go to Nvidia webiste here. Select latest TensorRT version that matches your CUDA version and download the DEB file.
sudo apt install ./<filename.deb>
sudo apt update
sudo apt install tensorrt
Verify that the trtexec utility is present.
whereis trtexec # should be trtexec: /usr/src/tensorrt/bin/trtexec
/usr/src/tensorrt/bin/trtexec
I used ...
sudo apt install nvidia-driver-470
However when I install the cuda toolkit later in the process the 470 ( recommended ) drivers are uninstalled ?????
The following packages will be REMOVED:
libnvidia-compute-470 libnvidia-compute-470:i386 libnvidia-decode-470
libnvidia-decode-470:i386 libnvidia-encode-470 libnvidia-encode-470:i386
nvidia-compute-utils-470 nvidia-driver-470 nvidia-utils-470
The following NEW packages will be installed:
libcuinj64-11.5 libnvidia-compute-495 libnvidia-compute-510
libnvidia-compute-525 libnvidia-ml-dev nsight-systems nsight-systems-target
nvidia-cuda-dev nvidia-cuda-toolkit nvidia-profiler nvidia-visual-profiler
This breaks my hardware support at runtime ( unsupported HW errors ).
Any clues ?