dvc.yaml

# Type dvc-help in this file and hit enter to get more information on how the extension can help to setup pipelines
stages:
  Features:
    cmd: python src/stages/features.py --config=params.yaml
    deps:
      - src/stages/features.py
    params:
      - features.fft_window_func
      - features.fft_window_size
      - features.fft_window_num
      - features.names
    outs:
      - Data/raw/dataset.csv

  Data: # delete ceratain level from training set
    cmd: python src/stages/del_load.py --config=params.yaml
    deps:
      - src/stages/del_load.py
    params:
      - data_split.load_exclude
      - features.fft_window_func
      - features.fft_window_size
      - features.fft_window_num
      - features.names
    outs:
      - Data/processed/X_train.csv
      - Data/processed/y_train.csv
      - Data/processed/X_test.csv
      - Data/processed/y_test.csv
      - Data/processed/load_train.csv
      - Data/processed/load_test.csv

  Training: # model training
    cmd: python src/stages/train.py --config=params.yaml
    deps:
      - src/stages/train.py
      - Data/processed/X_train.csv
      - Data/processed/y_train.csv
      - Data/processed/X_test.csv
      - Data/processed/y_test.csv
      - Data/processed/load_train.csv
      - Data/processed/load_test.csv
    params:
      - train.estimator_name
      - data_split.load_exclude
      - features.fft_window_func
      - features.fft_window_size
      - features.fft_window_num
      - features.names
    metrics:
      - reports/best_hp.json

  Evaluation: # evaluate model and save plots/reports
    always_changed: True
    cmd: python src/stages/evaluate.py --config=params.yaml
    deps:
      - src/stages/evaluate.py
      - Models/
      - Data/processed/X_test.csv
      - Data/processed/y_test.csv
    plots:
    - reports/confusion_matrix.png
    - reports/confusion_matrix_data.csv:
        template: confusion
        x: predicted
        y: y_true
    metrics:
      - reports/metrics.json