cifar10-job-template.yml

apiVersion: batch/v1
kind: Job
metadata:
  name: cifar10-single-job-$ITEM
spec:
  parallelism: 1
  template:
    metadata:
      name: cifar10-single-job
      labels:
        app: cifar10-single-job
    spec:
      restartPolicy: Never
      containers:
       - name: pytorch
         image: nvcr.io/nvidia/pytorch:18.11-py3
         workingDir: /cifar10-training
         env:
         - name: JOB_ID
           value: "$ITEM"
         command: ["bash"]
         args: ["-c","python cifar10_train.py"]
         computeResourceRequests: ["nvidia-gpu"]
         volumeMounts:
         - name: cifar10-training
           mountPath: /cifar10-training
         - name: cifar10-dataset
           mountPath: /datasets
      volumes:
      - name: cifar10-training
        gitRepo:
          repository: https://github.com/NVIDIA-developer-blog/kubernetes-hyperparam-exp.git
          revision: master
          directory: .
      - name: cifar10-dataset
        nfs:
          server: 10.110.114.22
          path: /home/sprasanna/Projects/training-datasets
      computeResources:
       - name: "nvidia-gpu"
         resources:
           limits:
             nvidia.com/gpu: 1