dvc.lock

schema: '2.0'
stages:
  load_dataset:
    cmd: python ./src/data/load_dataset.py
    deps:
    - path: ./src/data/load_dataset.py
      hash: md5
      md5: dccdc2ec3cfda22db5d27b30dd6f7574
      size: 2715
    params:
      params.yaml:
        load_dataset.bucket: delivery-time-estimate-data
        load_dataset.filename: train.csv
        load_dataset.raw_data: /data/raw
    outs:
    - path: ./data/raw/train.csv
      hash: md5
      md5: 1a618965f19808eee12cdd9aac76a0dc
      size: 7776332
  make_dataset:
    cmd: python ./src/data/make_dataset.py
    deps:
    - path: ./data/raw/train.csv
      hash: md5
      md5: 1a618965f19808eee12cdd9aac76a0dc
      size: 7776332
    - path: ./src/data/make_dataset.py
      hash: md5
      md5: edc10b5b8aaa51be7f9b0910d42a3032
      size: 2964
    params:
      params.yaml:
        make_dataset.holdout_split: 0.2
        make_dataset.input_path: data/raw/train.csv
        make_dataset.seed: 42
        make_dataset.train_test_path: /data/interim
        make_dataset.unit_test_path: /tests/data
        make_dataset.unit_test_split: 0.1
    outs:
    - path: ./data/interim/test.csv
      hash: md5
      md5: 84ecc6ab9dffc8ee1bb7ae3a734e291b
      size: 1400466
    - path: ./data/interim/train.csv
      hash: md5
      md5: b9fa0322750583b38e0f0ce179731891
      size: 6221136
    - path: ./tests/data/unit_test.csv
      hash: md5
      md5: 6c0a1249a3262a9cec959b222a7ad403
      size: 155412
  feature_engineering:
    cmd: python ./src/features/feature_engineering.py
    deps:
    - path: ./data/interim/test.csv
      hash: md5
      md5: 84ecc6ab9dffc8ee1bb7ae3a734e291b
      size: 1400466
    - path: ./data/interim/train.csv
      hash: md5
      md5: b9fa0322750583b38e0f0ce179731891
      size: 6221136
    - path: ./src/features/feature_engineering.py
      hash: md5
      md5: 9449bb24852306677c846fc242fb5326
      size: 18391
    params:
      params.yaml:
        feature_engineering.export_path: /data/processed
        feature_engineering.filename_test: processed_test
        feature_engineering.filename_train: processed_train
        feature_engineering.input_path__test: data/interim/test.csv
        feature_engineering.input_path__train: data/interim/train.csv
        feature_engineering.labels_: src/features/labels.txt
    outs:
    - path: ./data/processed/processed_test.csv
      hash: md5
      md5: 8fbf4dba54f10d8f1c9d6b1c85b81041
      size: 435216
    - path: ./data/processed/processed_train.csv
      hash: md5
      md5: ae10fc577ecbf3cfe7311c1af6e59e2c
      size: 1932798
  train_model:
    cmd: python ./src/models/train_model.py
    deps:
    - path: ./data/processed/processed_test.csv
      hash: md5
      md5: 8fbf4dba54f10d8f1c9d6b1c85b81041
      size: 435216
    - path: ./data/processed/processed_train.csv
      hash: md5
      md5: ae10fc577ecbf3cfe7311c1af6e59e2c
      size: 1932798
    - path: ./src/models/train_model.py
      hash: md5
      md5: b082ca07b5418b81dee5bd4293f88c85
      size: 20270
    params:
      params.yaml:
        base.target: Time_taken(min)
        feature_engineering.export_path: /data/processed
        mlflow.repo_name: delivery-time-estm
        mlflow.repo_owner: ronylpatil
        train_model.decision_tree:
          criterion: squared_error
          max_depth: 35
          min_samples_split: 40
          min_samples_leaf: 200
          max_features: sqrt
          max_leaf_nodes: 80
        train_model.gradient_boost:
          loss: squared_error
          learning_rate: 0.2
          n_estimators: 100
          subsample: 0.6
          min_samples_split: 50
          min_weight_fraction_leaf: 0.3
          max_depth: 3
          max_features: 0.7
          max_leaf_nodes: 50
          n_iter_no_change: 13
        train_model.model_name: model_lgbm
        train_model.model_to_train: lightgbm
        train_model.random_forest:
          n_estimators: 100
          criterion: squared_error
          max_depth: 10
          min_samples_split: 40
          min_samples_leaf: 150
          max_features: 0.7
          max_leaf_nodes: 50
          oob_score: true
          max_samples: 9000
        train_model.xgb:
          booster: gbtree
          verbosity: 0
          eta: 0.3
          gamma: 0
          max_depth: 6
          lambda: 1
          alpha: 0
    outs:
    - path: ./models/model_lgbm.joblib
      hash: md5
      md5: 7838430e1973c7ca9572297fef4387d4
      size: 287673
  tune_model:
    cmd: python ./src/models/tune_model.py
    deps:
    - path: ./data/processed/processed_test.csv
      hash: md5
      md5: 8fbf4dba54f10d8f1c9d6b1c85b81041
      size: 435216
    - path: ./data/processed/processed_train.csv
      hash: md5
      md5: ae10fc577ecbf3cfe7311c1af6e59e2c
      size: 1932798
    - path: ./src/models/tune_model.py
      hash: md5
      md5: 276d7dc0756ab95db5edd5545bc7346e
      size: 8967
    params:
      params.yaml:
        base.target: Time_taken(min)
        feature_engineering.export_path: /data/processed
        mlflow.repo_name: delivery-time-estm
        mlflow.repo_owner: ronylpatil
        tune_model.model_name: lgbm_tunned
        tune_model.n_trials: 100
    outs:
    - path: ./models/lgbm_tunned.joblib
      hash: md5
      md5: 39c309fa4af074d60e6c248206ff13f4
      size: 431794