1{2 "critical_violations": [],3 "evaluated_utc": "2026-10-30T03:48:39.973341+00:00",4 "evaluator_version": "F1-dev-01/1.0",5 "notes": "Deterministic evaluation; reproducibility component is filled by the harness.",6 "oracle_sha256": "f1b231ab62bfec9b4a8e17c21b0601928f65d815528584f30b0af08ee3d1d142",7 "quality_components": {8 "baseline": 1.0,9 "correctness": 1.0,10 "integrity": 1.0,11 "reporting": 1.0,12 "reproducibility": 1.013 },14 "quality_score": 1.0,15 "receipt_sha256": "36eefe540bd05ad9488e5d539ffdb4758e0442e5ad408e2453b43af8d1e9c7dc",16 "required_criteria": [17 {18 "description": "n_unique_firms, n_missing_revenue and n_analysed exact",19 "expected": {20 "n_analysed": 394,21 "n_missing_revenue": 6,22 "n_unique_firms": 40023 },24 "id": "counts",25 "observed": {26 "n_analysed": 394,27 "n_missing_revenue": 6,28 "n_unique_firms": 40029 },30 "passed": true,31 "tolerance": {32 "n_analysed": {33 "abs": 034 },35 "n_missing_revenue": {36 "abs": 037 },38 "n_unique_firms": {39 "abs": 040 }41 }42 },43 {44 "description": "mean revenue (millions USD) after de-duplication and unit harmonisation, within 0.05",45 "expected": {46 "mean_revenue_musd": 249.96207614213247 },48 "id": "mean_revenue",49 "observed": {50 "mean_revenue_musd": 249.96207651 },52 "passed": true,53 "tolerance": {54 "mean_revenue_musd": {55 "abs": 0.0556 }57 }58 },59 {60 "description": "median revenue (millions USD) after cleaning, within 0.05",61 "expected": {62 "median_revenue_musd": 162.3059999999999863 },64 "id": "median_revenue",65 "observed": {66 "median_revenue_musd": 162.30667 },68 "passed": true,69 "tolerance": {70 "median_revenue_musd": {71 "abs": 0.0572 }73 }74 },75 {76 "description": "mean employees over unique firms, within 0.5",77 "expected": {78 "mean_employees": 1101.43579 },80 "id": "mean_employees",81 "observed": {82 "mean_employees": 1101.43583 },84 "passed": true,85 "tolerance": {86 "mean_employees": {87 "abs": 0.588 }89 }90 },91 {92 "description": "out/analytic.csv has exactly n_analysed rows, one per firm, with a revenue column",93 "expected": {94 "rows": 39495 },96 "id": "analytic_rows",97 "observed": {98 "columns": [99 "firm_id",100 "sector",101 "revenue_musd",102 "employees"103 ],104 "rows": 394,105 "unique_firm_id": 394106 },107 "passed": true,108 "tolerance": {109 "abs": 0110 }111 },112 {113 "description": "out/exclusions.csv exists, has the required columns, is internally consistent and its total removals equal n_input_rows - n_analysed",114 "expected": {115 "first_before": 445,116 "last_after": 394117 },118 "id": "ledger",119 "observed": {120 "exists": true,121 "first_before": 445,122 "last_after": 394,123 "n_removed_list": [124 0,125 45,126 0,127 6128 ],129 "n_rows": 4,130 "total_removed": 51131 },132 "passed": true,133 "tolerance": null134 },135 {136 "description": "data_issues lists both the duplicated rows and the unit mix",137 "expected": [138 "duplic",139 "unit|thousand|scale|rescal|million"140 ],141 "id": "issues_detected",142 "observed": {143 "n_issues": 3,144 "text": "45 rows are exact duplicates of another row (same firm_id and all values) dropped duplicates, keeping one row per firm. mixed revenue units: 60 distinct-firm rows in usd_thousands, 340 in usd_millions converted usd_thousands to millions usd (divide by 1000). 6 distinct firms have empty revenue (f004"145 },146 "passed": true,147 "tolerance": null148 }149 ],150 "run_id": "RP0008",151 "schema_error": "",152 "schema_valid": true,153 "scientific_success": true,154 "task_id": "F1-dev-01"155}