spotty.yaml

project:
  name: calbert
  syncFilters:
    - exclude:
      - .git/*
      - .idea/*
      - '*/__pycache__/*'
      - .mypy_cache/*
      - export/*
      - run/*
      - docker/*
      - .vscode/*

container:
  projectDir: /workspace/project
  file: codegram/calbert:latest
  volumeMounts:
    - name: workspace
      mountPath: /workspace
  workingDir: /workspace/project
  commands: python -m calbert download_data --out-dir /workspace/data


instances:
  - name: i1
    provider: gcp
    parameters:
      zone: europe-west4-a
      onDemandInstance: false
      dockerDataRoot: /docker
      volumes:
        - name: workspace
          parameters:
            size: 100
            deletionPolicy: retain
        - name: docker
          parameters:
            size: 10
            mountDir: /docker
            deletionPolicy: retain 

      machineType: n1-standard-8
      gpu:
        type: nvidia-tesla-v100
        count: 1

scripts:
  test: python -m calbert train --tokenizer-path dist/tokenizer-uncased/ca.uncased.30000.model --train-path dist/data/train_subset.txt --valid-path dist/data/valid_subset.txt --train-batch-size 32 --eval-batch-size 32 --fp16 --export-path /workspace/test-export model=tiny
  train: python -m calbert train --tokenizer-path dist/tokenizer-uncased/ca.uncased.30000.model --train-path /workspace/data/train.txt --valid-path /workspace/data/valid.txt --train-batch-size 32 --eval-batch-size 32 --fp16 --export-path /workspace/export model=tiny