schema: '2.0'
stages:
  train:
    cmd: python src/models/train_model.py
    deps:
    - path: data/processed/train.csv
      md5: 51edd724b75a8e99a78b9138f8f37c60
      size: 25012573
    - path: data/processed/validation.csv
      md5: 0900e2bb330df94cb045faddd0b945d1
      size: 1138285
    - path: params.yml
      md5: d0f3e81bc9191e752a69761045a449d9
      size: 196
    - path: src/models/train_model.py
      md5: fca8acf70f09cecd679ca1ddb2eef6a9
      size: 1198
    outs:
    - path: models
      md5: 688745a9fb1cc7c8580887bae3873a39.dir
      size: 486952666
      nfiles: 10
    - path: reports/training_metrics.txt
      md5: 048a956b0eb431535d287bbc3322cf76
      size: 158
  eval:
    cmd: python src/models/evaluate_model.py
    deps:
    - path: data/processed/test.csv
      md5: 3cb7b63891f12d53b3ef3e81a2e93f8e
      size: 986944
    - path: models
      md5: 688745a9fb1cc7c8580887bae3873a39.dir
      size: 486952666
      nfiles: 10
    - path: params.yml
      md5: d0f3e81bc9191e752a69761045a449d9
      size: 196
    - path: src/models/evaluate_model.py
      md5: aa01b1564d737fef54ae45d25c5018d1
      size: 615
    outs:
    - path: reports/metrics.txt
      md5: 27d21366dca75caa1bb3777575cb126b
      size: 1596
  process_data:
    cmd: python src/data/process_data.py
    deps:
    - path: data/raw
      md5: d751713988987e9331980363e24189ce.dir
      size: 0
      nfiles: 0
    - path: params.yml
      md5: d0f3e81bc9191e752a69761045a449d9
      size: 196
    - path: src/data/process_data.py
      md5: ba3ba7b7c8a905b736b6b0a28d2334c4
      size: 623
    outs:
    - path: data/processed/test.csv
      md5: 3cb7b63891f12d53b3ef3e81a2e93f8e
      size: 986944
    - path: data/processed/train.csv
      md5: 51edd724b75a8e99a78b9138f8f37c60
      size: 25012573
    - path: data/processed/validation.csv
      md5: 0900e2bb330df94cb045faddd0b945d1
      size: 1138285