dvc.yaml

stages:
  preprocess_data:
    foreach:
      - train
      - val
      - test
    do:
      cmd: python scripts/preprocess_data.py --input-folder datasets/raw/${item} --output-folder datasets/preprocessed/${item}
      deps:
        - scripts/preprocess_data.py
        - datasets/raw/${item}
      outs:
        - datasets/preprocessed/${item}

  train_model:
    cmd: python scripts/train_model.py
    deps:
      - scripts/train_model.py
      - src/model.py
      - models/initial_weights.h5
      - datasets/preprocessed/train
      - datasets/preprocessed/val
    outs:
      - models/trained_model.h5

  compute_model_predictions:
    cmd: python scripts/compute_model_predictions.py
    deps:
      - scripts/compute_model_predictions.py
      - src/model.py
      - models/trained_model.h5
      - datasets/preprocessed/test
    outs:
      - predictions/test/model_predictions.csv

  match_predictions_with_ground_truths:
    ...

  compute_mean_average_precision:
    ...

  compute_confusion_matrix:
    ...