tutorials-and-examples/nvidia-bionemo/fine-tuning/job/job.yaml

apiVersion: batch/v1
kind: Job
metadata:
  name: esm2-finetuning
  namespace: bionemo-training
spec:
  backoffLimit: 3  # Number of retries before marking job as failed
  template:
    spec:
      containers:
      - name: finetuning
        image: nvcr.io/nvidia/clara/bionemo-framework:2.3
        command: ["python3"]
        args: ["/app/finetuning.py"]
        resources:
          limits:
            nvidia.com/gpu: 1
        volumeMounts:
        - name: bionemo-storage
          mountPath: /mnt/data
        - name: scripts
          mountPath: /app
      volumes:
      - name: bionemo-storage
        persistentVolumeClaim:
          claimName: bionemo-filestore
      - name: scripts
        configMap:
          name: finetuning-script
          defaultMode: 0755  
      restartPolicy: Never
      nodeSelector:
        cloud.google.com/gke-gpu: "true"