{
  "acquisition": {
    "acquired_at": "2026-06-27T22:25:04+00:00",
    "dedupe_fingerprint": "legacy:24713c7f7bfa78075849f7d2b4ac34626959d9e9f99930eeaf1a4ea0953ddaf9:openml-46931-363638-charges",
    "fetched_feature_columns": 6,
    "fetched_rows": 1338,
    "final_row_cap_policy": "strict_less_than_2000",
    "numeric_target_rows": 1338,
    "openml_data_id": 46931,
    "openml_name": "healthcare_insurance_expenses",
    "openml_target_feature": "charges",
    "openml_task_id": 363638,
    "openml_task_url": "https://www.openml.org/t/363638",
    "openml_url": "https://www.openml.org/d/46931",
    "row_policy": "kept_all",
    "source": "openml",
    "source_dataset_id": "46931",
    "source_dataset_name": "healthcare_insurance_expenses",
    "source_platform": "openml",
    "source_resource_id": "363638",
    "source_rows_after_target_cleanup": 1338,
    "source_target_column_original": "charges",
    "source_url": "https://www.openml.org/d/46931",
    "stage_max_rows": 50000,
    "stage_path": "/home/ubuntu/hyper/data/tabular/realreg-openml-46931-charges.csv",
    "stage_target_column": "charges",
    "staged_columns": 7,
    "staged_feature_columns": 6,
    "staged_rows": 1338,
    "staging_path": "/home/ubuntu/hyper/data/tabular/realreg-openml-46931-charges.csv",
    "target_feature": "charges",
    "target_max": 63770.42801,
    "target_min": 1121.8739,
    "target_unique_values": 1337,
    "tool": "models.tools.acquire_real_regression_datasets"
  },
  "counts": {
    "dropped_feature_count": 0,
    "n_features": 6,
    "original_columns": 7,
    "original_rows": 1338,
    "rows_after_target_cleanup": 1338,
    "selected_rows": 1338,
    "strata_count": 99,
    "test_rows": 338,
    "train_rows": 1000
  },
  "created_at": "2026-06-27T22:25:04+00:00",
  "dataset_hash": "24713c7f7bfa7807",
  "dedupe_fingerprint": "legacy:24713c7f7bfa78075849f7d2b4ac34626959d9e9f99930eeaf1a4ea0953ddaf9:openml-46931-363638-charges",
  "dropped_feature_columns": [],
  "feature_columns": [
    "age",
    "sex",
    "bmi",
    "children",
    "smoker",
    "region"
  ],
  "feature_encodings": {
    "age": {
      "impute_value": 39.0,
      "kind": "numeric"
    },
    "bmi": {
      "impute_value": 30.4,
      "kind": "numeric"
    },
    "children": {
      "impute_value": 1.0,
      "kind": "numeric"
    },
    "region": {
      "impute_value": 2.0,
      "kind": "categorical_codes",
      "mapping": {
        "northeast": 0,
        "northwest": 1,
        "southeast": 2,
        "southwest": 3
      }
    },
    "sex": {
      "impute_value": 1.0,
      "kind": "categorical_codes",
      "mapping": {
        "female": 0,
        "male": 1
      }
    },
    "smoker": {
      "impute_value": 0.0,
      "kind": "categorical_codes",
      "mapping": {
        "no": 0,
        "yes": 1
      }
    }
  },
  "feature_scaling": {
    "constant_columns": [],
    "max": [
      64.0,
      1.0,
      53.13,
      5.0,
      1.0,
      3.0
    ],
    "method": "minmax over cleaned source rows before sampling",
    "min": [
      18.0,
      0.0,
      15.96,
      0.0,
      0.0,
      0.0
    ],
    "range": [
      0.0,
      1.0
    ]
  },
  "hash_algorithm": "sha256:first16",
  "headers_removed_from_csv": true,
  "n_features": 6,
  "origin": "real",
  "original_headers": [
    "age",
    "sex",
    "bmi",
    "children",
    "smoker",
    "region",
    "charges"
  ],
  "output_headers": [
    "age",
    "sex",
    "bmi",
    "children",
    "smoker",
    "region",
    "charges"
  ],
  "retrieved_at": "2026-06-27T22:25:04+00:00",
  "sampling": {
    "final_max_rows": 1999,
    "method": "quota stratified by target distribution plus high-variance feature quantile bins",
    "minimum_final_rows": 101,
    "minimum_test_to_train_ratio_when_small": 0.2,
    "seed": 625220466,
    "source_read_max_rows": null,
    "test_target_rows": 999,
    "train_target_rows": 1000
  },
  "sha256": "24713c7f7bfa78075849f7d2b4ac34626959d9e9f99930eeaf1a4ea0953ddaf9",
  "source": "openml",
  "source_dataset_id": "46931",
  "source_dataset_name": "healthcare_insurance_expenses",
  "source_file": "realreg-openml-46931-charges.csv",
  "source_license": null,
  "source_modified_at": "2026-06-27T22:25:04+00:00",
  "source_openml_data_id": 46931,
  "source_openml_name": "healthcare_insurance_expenses",
  "source_openml_task_id": 363638,
  "source_path": "/home/ubuntu/hyper/data/tabular/realreg-openml-46931-charges.csv",
  "source_platform": "openml",
  "source_real_data": true,
  "source_resource_id": "363638",
  "source_rows": 1338,
  "source_size_bytes": 54351,
  "source_target_column_original": "charges",
  "source_target_key": "openml:46931::charges",
  "source_task_url": "https://www.openml.org/t/363638",
  "source_url": "https://www.openml.org/d/46931",
  "staged_rows": 1338,
  "staging_path": "/home/ubuntu/hyper/data/tabular/realreg-openml-46931-charges.csv",
  "target_column": "charges",
  "target_generation": "original_numeric_column",
  "target_is_final_column": true,
  "target_names": [
    "charges"
  ],
  "target_scaling": {
    "max": 63770.42801,
    "min": 1121.8739,
    "mode": "regression",
    "range": [
      0.0,
      1.0
    ]
  },
  "target_source_encoding": {
    "kind": "numeric"
  },
  "task_type": "regression",
  "test_rows": 338,
  "train_rows": 1000
}