AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |
Back to Blog
How to install curl tf23/29/2023 ![]() $ docker run -gpus '"device=0"' -it -rm -v $(pwd):/work -w /workspace/tensorrt/data/resnet50/ nvcr.io/nvidia/tensorrt:20.11-p圓 * We only have to pull docker image only one time ![]() ![]() # docker pull nvcr.io/nvidia/tensorrt:20.11-p圓 # download docker and do benchmark(tensorrt) If it's OK, it may show message like picture below: # docker run -gpus all nvidia/cuda:9.0-base nvidia-smi $ sudo apt-get install -y nvidia-container-toolkit $ curl -s -L $distribution/nvidia-docker.list | sudo tee /etc/apt//nvidia-docker.list $ sudo apt-get install docker-ce docker-ce-cli containerd.io $ sudo apt-get remove docker docker-engine docker.io containerd runcĩDC8 5822 9FC7 DD38 854A E2D8 8D81 803C 0EBF CD88 If you install the driver successfully, it shows message like picture below: run file, click "continue install" option in every error message NVIDIA-Linux-x86_64-455.38.run //notice the filename when you click command # chmod 777 NVIDIA-Linux-x86_64-455.38.run //notice the filename when you click command, filename may be different because of its version or others Sudo systemctl set-default runlevel3.targetĭownload the latest stable Driver from NVIDIA offical website. ![]() Sudo systemctl set-default multi-user.target I installed Ubuntu 18.04 Desktop version because changing to CLI interface is easier than upgrading linux kernel OS: Ubuntu 18.04 LTS Desktop, kernel 5.4.0 (UEFI) Mother Board: BIS-3101 with NVIDIA GEFORCE RTX-3070/3080 x 1ĬPU: Intel® Core® i7 9700E CPU 2.60GHz x 1 # tags: NVIDIA GEFORCE RTX-3070, NVIDIA GEFORCE RTX-3080 # NVIDIA RTX-3070/3080 install and benchmark guide Tags: NVIDIA, NVIDIA GEFORCE RTX-3070, NVIDIA GEFORCE RTX-3080 ![]()
0 Comments
Read More
Leave a Reply. |