{
  "acquisition": {
    "acquired_at": "2026-06-26T09:01:31+00:00",
    "fetched_feature_columns": 16,
    "fetched_rows": 1340,
    "openml_data_id": 966,
    "openml_name": "analcatdata_halloffame",
    "openml_url": "https://www.openml.org/d/966",
    "row_policy": "kept_all",
    "source": "openml",
    "source_rows_after_target_cleanup": 1340,
    "stage_path": "/home/ubuntu/hyper/data/tabular/analcatdata-halloffame.csv",
    "staged_feature_columns": 16,
    "staged_rows": 1340,
    "tool": "models.tools.acquire_real_binary_datasets"
  },
  "counts": {
    "dropped_feature_count": 0,
    "n_features": 16,
    "original_columns": 17,
    "original_rows": 1340,
    "rows_after_target_cleanup": 1340,
    "selected_rows": 1340,
    "strata_count": 78,
    "test_rows": 340,
    "train_rows": 1000
  },
  "created_at": "2026-06-26T09:01:31+00:00",
  "dataset_hash": "a5110bc75977b78a",
  "dropped_feature_columns": [],
  "feature_columns": [
    "Number_seasons",
    "Games_played",
    "At_bats",
    "Runs",
    "Hits",
    "Doubles",
    "Triples",
    "Home_runs",
    "RBIs",
    "Walks",
    "Strikeouts",
    "Batting_average",
    "On_base_pct",
    "Slugging_pct",
    "Fielding_ave",
    "Position"
  ],
  "feature_encodings": {
    "At_bats": {
      "impute_value": 4302.0,
      "kind": "numeric"
    },
    "Batting_average": {
      "impute_value": 0.267,
      "kind": "numeric"
    },
    "Doubles": {
      "impute_value": 184.5,
      "kind": "numeric"
    },
    "Fielding_ave": {
      "impute_value": 0.973,
      "kind": "numeric"
    },
    "Games_played": {
      "impute_value": 1282.5,
      "kind": "numeric"
    },
    "Hits": {
      "impute_value": 1168.0,
      "kind": "numeric"
    },
    "Home_runs": {
      "impute_value": 51.0,
      "kind": "numeric"
    },
    "Number_seasons": {
      "impute_value": 13.0,
      "kind": "numeric"
    },
    "On_base_pct": {
      "impute_value": 0.335,
      "kind": "numeric"
    },
    "Position": {
      "impute_value": 3.0,
      "kind": "numeric"
    },
    "RBIs": {
      "impute_value": 486.0,
      "kind": "numeric"
    },
    "Runs": {
      "impute_value": 575.0,
      "kind": "numeric"
    },
    "Slugging_pct": {
      "impute_value": 0.38,
      "kind": "numeric"
    },
    "Strikeouts": {
      "impute_value": 365.5,
      "kind": "numeric"
    },
    "Triples": {
      "impute_value": 40.0,
      "kind": "numeric"
    },
    "Walks": {
      "impute_value": 380.0,
      "kind": "numeric"
    }
  },
  "feature_scaling": {
    "constant_columns": [],
    "max": [
      26.0,
      3562.0,
      14053.0,
      2246.0,
      4256.0,
      792.0,
      309.0,
      755.0,
      2297.0,
      2056.0,
      2597.0,
      0.366,
      0.483,
      0.69,
      1.0,
      6.0
    ],
    "method": "minmax over cleaned source rows before sampling",
    "min": [
      10.0,
      140.0,
      252.0,
      20.0,
      48.0,
      6.0,
      0.0,
      0.0,
      21.0,
      17.0,
      0.0,
      0.161,
      0.194,
      0.201,
      0.82,
      0.0
    ],
    "range": [
      0.0,
      1.0
    ]
  },
  "hash_algorithm": "sha256:first16",
  "headers_removed_from_csv": true,
  "n_features": 16,
  "origin": "real",
  "original_headers": [
    "Number_seasons",
    "Games_played",
    "At_bats",
    "Runs",
    "Hits",
    "Doubles",
    "Triples",
    "Home_runs",
    "RBIs",
    "Walks",
    "Strikeouts",
    "Batting_average",
    "On_base_pct",
    "Slugging_pct",
    "Fielding_ave",
    "Position",
    "target"
  ],
  "output_headers": [
    "Number_seasons",
    "Games_played",
    "At_bats",
    "Runs",
    "Hits",
    "Doubles",
    "Triples",
    "Home_runs",
    "RBIs",
    "Walks",
    "Strikeouts",
    "Batting_average",
    "On_base_pct",
    "Slugging_pct",
    "Fielding_ave",
    "Position",
    "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": 2753834186,
    "source_read_max_rows": null,
    "test_target_rows": 1000,
    "train_target_rows": 1000
  },
  "sha256": "a5110bc75977b78a584534191097912da61be721a6a240b8272ccd1474ccfb2e",
  "source": "openml",
  "source_file": "analcatdata-halloffame.csv",
  "source_modified_at": "2026-06-26T09:01:31+00:00",
  "source_openml_data_id": 966,
  "source_openml_name": "analcatdata_halloffame",
  "source_path": "/home/ubuntu/hyper/data/tabular/analcatdata-halloffame.csv",
  "source_real_data": true,
  "source_size_bytes": 130180,
  "source_url": "https://www.openml.org/d/966",
  "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": 340,
  "train_rows": 1000
}