job-template.yaml

apiVersion: batch/v1
kind: Job
metadata:
  name: process-item-$ITEM
  labels:
    jobgroup: pretrain
spec:
  template:
    metadata:
      name: pretrain-job
      labels:
        jobgroup: pretrain
    spec:
      containers:
      - name: c
        image: gcr.io/blueface/bert
        env:
          - name: "GOOGLE_APPLICATION_CREDENTIALS"
            value: "/var/run/secret/cloud.google.com/service-account.json"
        command: ["/usr/bin/python3", "create_pretraining_data.py",
                  "--input_file=gs://koursaros/enwiki-feb-doc-split/wiki_$ITEM.txt.doc.train",
                  "--output_file=gs://koursaros/bert_pretrain_data/wiki_$ITEM.tf_record",
                  "--vocab_file=vocab.txt"]
        resources:
          requests:
            memory: "3Gi"
        volumeMounts:
          - name: "service-account"
            mountPath: "/var/run/secret/cloud.google.com"
      volumes:
        - name: "service-account"
          secret:
            secretName: "my-app-sa-key"
      restartPolicy: Never