schema: '2.0'
stages:
  download_dataset@enron_spam:
    cmd: PYTHONPATH=. python experiments/scripts/download_dataset.py --dataset_name
      enron_spam --output_dir data/datasets/enron_spam
    deps:
    - path: experiments/scripts/download_dataset.py
      md5: 9eb915fd5b9216965db519f686408a51
      size: 887
    outs:
    - path: data/datasets/enron_spam/
      md5: 66d44efedf37990b1989c81bbee085e0.dir
      size: 53096069
      nfiles: 3
  get_model@enron_spam:
    cmd: PYTHONPATH=. python experiments/scripts/get_model.py --dataset_name enron_spam
      --output_dir data/models/enron_spam
    deps:
    - path: data/datasets/enron_spam
      md5: 66d44efedf37990b1989c81bbee085e0.dir
      size: 53096069
      nfiles: 3
    - path: experiments/scripts/get_model.py
      md5: 5050f51b4019bba97af47971f6c7cab4
      size: 747
    outs:
    - path: data/models/enron_spam/
      md5: 3e16b22f59532c66beeadea958e0579a.dir
      size: 18505614
      nfiles: 6
  classify@enron_spam:
    cmd: PYTHONPATH=. python experiments/scripts/classify.py --dataset_name enron_spam
      --output_dir data/classification/enron_spam
    deps:
    - path: data/datasets/enron_spam/
      md5: 66d44efedf37990b1989c81bbee085e0.dir
      size: 53096069
      nfiles: 3
    - path: data/models/enron_spam/
      md5: 3e16b22f59532c66beeadea958e0579a.dir
      size: 18505614
      nfiles: 6
    - path: experiments/scripts/classify.py
      md5: 50f55b90eb47cbf448d83f8392dd37b6
      size: 1102
    outs:
    - path: data/classification/enron_spam
      md5: c7d42825b98b289f6a5ed3be1af14413.dir
      size: 2763843
      nfiles: 2
  explain@enron_spam:
    cmd: PYTHONPATH=. python experiments/scripts/explain.py --dataset_name enron_spam
      --output_dir data/explanations/enron_spam
    deps:
    - path: data/datasets/enron_spam
      md5: 66d44efedf37990b1989c81bbee085e0.dir
      size: 53096069
      nfiles: 3
    - path: data/models/enron_spam
      md5: 3e16b22f59532c66beeadea958e0579a.dir
      size: 18505614
      nfiles: 6
    - path: experiments/scripts/explain.py
      md5: c85cbb774f2682ee39948e701fa0b0ca
      size: 1445
    outs:
    - path: data/explanations/enron_spam/
      md5: 376bd1619c08b4989564788e74de8e06.dir
      size: 7870394
      nfiles: 1
  download_dataset@poleval:
    cmd: PYTHONPATH=. python experiments/scripts/download_dataset.py --dataset_name
      poleval --output_dir data/datasets/poleval
    deps:
    - path: experiments/scripts/download_dataset.py
      md5: 9eb915fd5b9216965db519f686408a51
      size: 887
    outs:
    - path: data/datasets/poleval/
      md5: 826f974f794e24efcb5aedb054d1fd55.dir
      size: 1688836
      nfiles: 3
  preprocess_dataset@poleval:
    cmd: PYTHONPATH=. python experiments/scripts/tag_dataset.py --dataset_name poleval
    deps:
    - path: data/datasets/poleval/
      md5: 826f974f794e24efcb5aedb054d1fd55.dir
      size: 1688836
      nfiles: 3
    - path: experiments/scripts/tag_dataset.py
      md5: 1d911edcd336cacaec482e6b7570eb1a
      size: 2716
    outs:
    - path: data/preprocessed/poleval/
      md5: 8daba6ad0597214499ac9b96e8e47c9f.dir
      size: 501920
      nfiles: 1
  preprocess_dataset@enron_spam:
    cmd: PYTHONPATH=. python experiments/scripts/tag_dataset.py --dataset_name enron_spam
    deps:
    - path: data/datasets/enron_spam/
      md5: 66d44efedf37990b1989c81bbee085e0.dir
      size: 53096069
      nfiles: 3
    - path: experiments/scripts/tag_dataset.py
      md5: 1d911edcd336cacaec482e6b7570eb1a
      size: 2716
    outs:
    - path: data/preprocessed/enron_spam/
      md5: 80c8dd3aa3bacf3afe8cf3138ab01d00.dir
      size: 10639521
      nfiles: 1