Spaces:
Sleeping
Sleeping
File size: 2,550 Bytes
4a18208 2d793d9 4a18208 2d793d9 4a18208 d957033 bcb678f d957033 bcb678f 2d793d9 d957033 4a18208 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 | {
"easy": {
"task_id": "easy_dedup_rename",
"difficulty": "easy",
"description": "Clean an employee dataset by: 1) Removing duplicate rows, 2) Renaming columns to snake_case format. Call finish when done.",
"max_steps": 10,
"operations_allowed": [
"remove_duplicates",
"rename_columns",
"finish"
],
"scoring": {
"duplicate_score": 0.5,
"schema_score": 0.5
}
},
"medium": {
"task_id": "medium_missing_dtype",
"difficulty": "medium",
"description": "Clean a customer dataset by: 1) Filling missing values using correct strategy (mean for numeric, mode for categorical), 2) Fixing data types (age, purchases, salary must be numeric). Call finish when done.",
"max_steps": 15,
"operations_allowed": [
"fill_missing_mean",
"fill_missing_mode",
"fill_missing_median",
"fix_dtype",
"finish"
],
"scoring": {
"missing_score": 0.5,
"dtype_score": 0.5
}
},
"hard": {
"task_id": "hard_full_pipeline",
"difficulty": "hard",
"description": "Clean an orders dataset by running a full pipeline: 1) Remove duplicate orders, 2) Fill missing values, 3) Fix data types, 4) Remove outliers in quantity and price columns, 5) Validate final schema. Call finish when done.",
"max_steps": 20,
"operations_allowed": [
"remove_duplicates",
"fill_missing_mean",
"fill_missing_mode",
"fill_missing_median",
"fix_dtype",
"remove_outliers",
"validate_schema",
"finish"
],
"scoring": {
"duplicate_score": 0.2,
"missing_score": 0.2,
"dtype_score": 0.2,
"outlier_score": 0.2,
"schema_score": 0.2
}
},
"expert": {
"task_id": "expert_sales_pipeline",
"difficulty": "expert",
"description": "Clean a sales dataset by running expert pipeline in ORDER: 1) Remove duplicate transactions, 2) Rename columns to snake_case, 3) Fill missing values using mode strategy, 4) Fix all data types, 5) Remove outliers using IQR method, 6) Validate final schema. Call finish when done.",
"max_steps": 25,
"operations_allowed": [
"remove_duplicates",
"rename_columns",
"fill_missing_mean",
"fill_missing_mode",
"fill_missing_median",
"fix_dtype",
"remove_outliers",
"validate_schema",
"finish"
],
"scoring": {
"duplicate_score": 0.15,
"missing_score": 0.2,
"dtype_score": 0.2,
"outlier_score": 0.2,
"schema_score": 0.25
}
}
} |