Search This Blog

3/18/19

How to use PyTorch on QNAP NAS with Container Station?

How to use PyTorch with Container Station?

What is PyTorch?

It’s a Python-based scientific computing package targeted at two sets of audiences:
  • A replacement for NumPy to use the power of GPUs
  • A deep learning research platform that provides maximum flexibility and speed

Installation Instructions

Recommended version

QTS 4.3.5/4.3.6 and Nvidia Driver qpkg v1.3.5
Tag
Pull command
Description
pytorch/pytorch:0.4.1-cuda9-cudnn7-devel
docker pull pytorch/pytorch:0.4.1-cuda9-cudnn7-devel



QTS 4.4.x and Nvidia Driver qpkg v2.0.0
Tag
Pull command
Description
pytorch/pytorch:1.0.1-cuda10.0-cudnn7-devel
docker pull pytorch/pytorch:1.0.1-cuda10.0-cudnn7-devel


Before running the PyTorch container, use docker pull command or click the Pull Button to ensure the desired image is installed. Once the pull is complete, you can go to next step.




Go to Control Panel -> Hardware -> Graphics Card.  Assign the GPU to the Container Station for Resource Use.



Go to Container Station > Click "Create". Search  "PyTorch " and click  "Install/Create".





Using Command to mount NVIDIA GPU



Using SSH to NAS and then can follow the below steps:


Add below commands when running Docker Run:
--device /dev/nvidia0:/dev/nvidia0 \
--device /dev/nvidiactl:/dev/nvidiactl \
--device /dev/nvidia-uvm:/dev/nvidia-uvm \

-v `/sbin/getcfg NVIDIA_GPU_DRV Install_Path -f /etc/config/qpkg.conf -d None`/usr/:/usr/local/nvidia


related information:


No comments:

Post a Comment

Related Posts Plugin for WordPress, Blogger...