• for IDA notebook job apply: make sure the name in the metadata are the same in the notebookGPU.yaml, routes and services yaml files are the same.

• jxcodetw/jupyter-pytorch
• inferislux/zaiqiao-ml:latest

Three types of node are avaliable:

• gpu2080ti
• gpu3090
• gputitan
• gpua6000

specify the node type in the yaml file, e.g. `nodeselector: node-role.ida/gpu2080ti: 'true'

• basic yaml file for requesting a gpu node

Needed files: pythonmlnotebookgpu.yaml & routes.yaml & service.yaml

• If you need to mount on multiple volclaim

Needed files: pythonmlnotebookgpudenseqe.yaml & routes.yaml & service.yaml

• If you need to request a second gpu node

Needed files: pythonmlnotebook3.yaml & pythonmlnotebook3routes.yaml & pythonmlnotebook3service.yaml