model-build.yaml

apiVersion: batch/v1
kind: Job
metadata:
  name: conv-subhasis007
spec:
  template:
    spec:
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
              - matchExpressions:
                  - key: nvidia.com/gpu.product
                    operator: In
                    values:
                      - NVIDIA-A10
                      - NVIDIA-A100-SXM4-80GB

      tolerations:
         - effect: NoSchedule
           key: nautilus.io/nrp-testing
           operator: Exists
      containers:
        - name: intent
          image: subhasis79/conv_intent_nourish:latest
          envFrom:
            - secretRef:
                name: subhasis-env
          command: ["/bin/sh", "-c"]
          args:
            - "python code/conv_intent.py; huggingface-cli login --token $HUG_TOKEN_RT; sleep 18000;"
          resources:
            limits:
              memory: 32Gi
              cpu: 20m
              nvidia.com/gpu: 1
            requests:
              memory: 32Gi
              cpu: 20m
              nvidia.com/gpu: 1
      restartPolicy: Never
  backoffLimit: 3