dvc.lock

schema: '2.0'
stages:
  data_ingestion:
    cmd: python -m ml.comment_sentiment.ingestion
    deps:
    - path: ml/comment_sentiment/ingestion.py
      hash: md5
      md5: c46bda70cd7013476c8022cf85d694fd
      size: 2509
    params:
      params.yaml:
        dataset:
          url:
            https://raw.githubusercontent.com/Himanshu-1703/reddit-sentiment-analysis/refs/heads/main/data/reddit.csv
          rename_columns:
            clean_comment: text
            category: target
          train_size: 0.75
          target_labels:
            0: neg
            1: neu
            2: pos
        ingestion:
          processed_train_path: data/processed/train.parquet
          processed_test_path: data/processed/test.parquet
    outs:
    - path: data/processed/test.parquet
      hash: md5
      md5: b59e6dbce031d40ae776b7540775d3d2
      size: 651415
    - path: data/processed/train.parquet
      hash: md5
      md5: d0f05f0713cd9c610574593ac8f67d58
      size: 1966155
  model_building:
    cmd: python -m ml.comment_sentiment.building
    deps:
    - path: data/processed/train.parquet
      hash: md5
      md5: d0f05f0713cd9c610574593ac8f67d58
      size: 1966155
    - path: ml/comment_sentiment/building.py
      hash: md5
      md5: f16b1f3812b97a3cab80948cd048a5dd
      size: 2659
    params:
      params.yaml:
        model:
          module: sklearn.ensemble
          name: HistGradientBoostingClassifier
          params: {}
        vectorizer:
          module: sklearn.feature_extraction.text
          name: TfidfVectorizer
          params:
            max_features: 5000
    outs:
    - path: models/classifier.pkl
      hash: md5
      md5: bc1e6448aa94bb576f3017c5796ed2fe
      size: 3690247
  model_evaluation:
    cmd: python -m ml.comment_sentiment.evaluation
    deps:
    - path: data/processed/test.parquet
      hash: md5
      md5: b59e6dbce031d40ae776b7540775d3d2
      size: 651415
    - path: ml/comment_sentiment/evaluation.py
      hash: md5
      md5: a365f24a1738ac4b01a644642071595e
      size: 5113
    - path: models/classifier.pkl
      hash: md5
      md5: bc1e6448aa94bb576f3017c5796ed2fe
      size: 3690247
    params:
      params.yaml:
        evaluation:
          train_vec_path: models/train_vec_data.pkl
        mlflow:
          tracking_uri: ./mlruns
          experiment_name: exp1-arv-testing
    outs:
    - path: mlflow_run_info.json
      hash: md5
      md5: 4e8a3b62c18edc0906f89907a2703b1b
      size: 75