{
  "acquisition": {
    "acquired_at": "2026-06-26T09:16:56+00:00",
    "fetched_feature_columns": 10,
    "fetched_rows": 10000,
    "openml_data_id": 46911,
    "openml_name": "Bank_Customer_Churn",
    "openml_url": "https://www.openml.org/d/46911",
    "row_policy": "stratified_cap_to_final_max",
    "source": "openml",
    "source_rows_after_target_cleanup": 10000,
    "stage_path": "/home/ubuntu/hyper/data/tabular/bank-customer-churn.csv",
    "staged_feature_columns": 10,
    "staged_rows": 2000,
    "tool": "models.tools.acquire_real_binary_datasets"
  },
  "counts": {
    "dropped_feature_count": 0,
    "n_features": 10,
    "original_columns": 11,
    "original_rows": 2000,
    "rows_after_target_cleanup": 2000,
    "selected_rows": 2000,
    "strata_count": 16,
    "test_rows": 1000,
    "train_rows": 1000
  },
  "created_at": "2026-06-26T09:16:56+00:00",
  "dataset_hash": "7bd4c609c8fc6af5",
  "dropped_feature_columns": [],
  "feature_columns": [
    "credit_score",
    "country",
    "gender",
    "age",
    "tenure",
    "balance",
    "products_number",
    "credit_card",
    "active_member",
    "estimated_salary"
  ],
  "feature_encodings": {
    "active_member": {
      "impute_value": 1.0,
      "kind": "numeric"
    },
    "age": {
      "impute_value": 37.0,
      "kind": "numeric"
    },
    "balance": {
      "impute_value": 98998.63,
      "kind": "numeric"
    },
    "country": {
      "impute_value": 0.5,
      "kind": "numeric"
    },
    "credit_card": {
      "impute_value": 1.0,
      "kind": "numeric"
    },
    "credit_score": {
      "impute_value": 651.0,
      "kind": "numeric"
    },
    "estimated_salary": {
      "impute_value": 102823.885,
      "kind": "numeric"
    },
    "gender": {
      "impute_value": 1.0,
      "kind": "numeric"
    },
    "products_number": {
      "impute_value": 1.0,
      "kind": "numeric"
    },
    "tenure": {
      "impute_value": 5.0,
      "kind": "numeric"
    }
  },
  "feature_scaling": {
    "constant_columns": [],
    "max": [
      850.0,
      2.0,
      1.0,
      79.0,
      10.0,
      221532.8,
      4.0,
      1.0,
      1.0,
      199970.74
    ],
    "method": "minmax over cleaned source rows before sampling",
    "min": [
      358.0,
      0.0,
      0.0,
      18.0,
      0.0,
      0.0,
      1.0,
      0.0,
      0.0,
      90.07
    ],
    "range": [
      0.0,
      1.0
    ]
  },
  "hash_algorithm": "sha256:first16",
  "headers_removed_from_csv": true,
  "n_features": 10,
  "origin": "real",
  "original_headers": [
    "credit_score",
    "country",
    "gender",
    "age",
    "tenure",
    "balance",
    "products_number",
    "credit_card",
    "active_member",
    "estimated_salary",
    "target"
  ],
  "output_headers": [
    "credit_score",
    "country",
    "gender",
    "age",
    "tenure",
    "balance",
    "products_number",
    "credit_card",
    "active_member",
    "estimated_salary",
    "target"
  ],
  "sampling": {
    "final_max_rows": 2000,
    "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": 2061623556,
    "source_read_max_rows": null,
    "test_target_rows": 1000,
    "train_target_rows": 1000
  },
  "sha256": "7bd4c609c8fc6af58249988142aa8fab28f6e356b037a389fbb39b16c8bdd0d0",
  "source": "openml",
  "source_file": "bank-customer-churn.csv",
  "source_modified_at": "2026-06-26T09:16:56+00:00",
  "source_openml_data_id": 46911,
  "source_openml_name": "Bank_Customer_Churn",
  "source_path": "/home/ubuntu/hyper/data/tabular/bank-customer-churn.csv",
  "source_real_data": true,
  "source_size_bytes": 112141,
  "source_url": "https://www.openml.org/d/46911",
  "target_column": "target",
  "target_is_final_column": true,
  "target_names": [
    "target"
  ],
  "target_scaling": {
    "class_to_code": {
      "0.0": 0,
      "1.0": 1
    },
    "classes": [
      0.0,
      1.0
    ],
    "mode": "classification",
    "n_classes": 2,
    "range": [
      0.0,
      1.0
    ],
    "scaled_code_formula": "code / max(1, n_classes - 1)"
  },
  "target_source_encoding": {
    "kind": "numeric"
  },
  "task_type": "binary",
  "test_rows": 1000,
  "train_rows": 1000
}