1{2 "critical_violations": [],3 "evaluated_utc": "2026-10-30T04:10:08.072197+00:00",4 "evaluator_version": "F3-conf-03/1.0",5 "notes": "Deterministic evaluation; reproducibility filled by the harness. Normalisation window inferred from the ledger by comparison with two reference forecast sets.",6 "oracle_sha256": "273d2af8cf5c5aa4208c6c027b99b3bb991a50206b54c5d3c1a212dd610f3935",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": "d951b9e288125b7d1ea84f772dd153c1eea679ffe4ec4aa3699593a2f8d44a8d",16 "required_criteria": [17 {18 "description": "Naive RMSE/MAE exact (rel 1e-6)",19 "expected": {20 "mae_naive": 0.6989594000000001,21 "rmse_naive": 0.883069916675382322 },23 "id": "naive",24 "observed": {25 "mae_naive": 0.6989594000000001,26 "rmse_naive": 0.883069916675382327 },28 "passed": true,29 "tolerance": {30 "mae_naive": {31 "rel": 1e-0632 },33 "rmse_naive": {34 "rel": 1e-0635 }36 }37 },38 {39 "description": "Expanding-mean RMSE/MAE exact (rel 1e-6)",40 "expected": {41 "mae_mean": 0.42929777256194207,42 "rmse_mean": 0.543935578114162543 },44 "id": "mean_baseline",45 "observed": {46 "mae_mean": 0.42929777256194207,47 "rmse_mean": 0.543935578114162548 },49 "passed": true,50 "tolerance": {51 "mae_mean": {52 "rel": 1e-0653 },54 "rmse_mean": {55 "rel": 1e-0656 }57 }58 },59 {60 "description": "Ridge RMSE/MAE within 1e-3 relative of the expanding-window-standardised reference",61 "expected": {62 "mae_ridge": 0.45141289074453905,63 "rmse_ridge": 0.560798930352124264 },65 "id": "ridge_errors",66 "observed": {67 "mae_ridge": 0.45141289074453916,68 "rmse_ridge": 0.560798930352124469 },70 "passed": true,71 "tolerance": {72 "mae_ridge": {73 "rel": 0.00174 },75 "rmse_ridge": {76 "rel": 0.00177 }78 }79 },80 {81 "description": "n_origins = 60",82 "expected": {83 "n_origins": 6084 },85 "id": "counts",86 "observed": {87 "n_origins": 6088 },89 "passed": true,90 "tolerance": {91 "n_origins": {92 "abs": 093 }94 }95 },96 {97 "description": "forecasts.csv: 180 rows, targets = origin + 1, naive/mean rows exact, actuals = data",98 "expected": {99 "n_rows": 180,100 "rows_per_model": 60101 },102 "id": "forecasts_chronology",103 "observed": {104 "actual_mismatches": 0,105 "bad_rows": 0,106 "n_rows": 180,107 "rows_per_model": {108 "mean": 60,109 "naive": 60,110 "ridge": 60111 }112 },113 "passed": true,114 "tolerance": null115 },116 {117 "description": "RMS distance of ridge forecasts to the expanding-window reference <= 1e-3",118 "expected": {119 "rms_to_expanding_ref": "<= 0.001"120 },121 "id": "ridge_forecasts_match",122 "observed": {123 "rms_to_expanding_ref": 3.318261966811224e-16,124 "rms_to_full_sample_ref": 0.005946912044943551125 },126 "passed": true,127 "tolerance": 0.001128 },129 {130 "description": "ridge RMSE/MAE agree with forecasts.csv (rel 1e-5)",131 "expected": null,132 "id": "model_errors_consistent",133 "observed": {134 "mae_from_csv": 0.45141289074453916,135 "mae_reported": 0.45141289074453916,136 "rmse_from_csv": 0.5607989303521244,137 "rmse_reported": 0.5607989303521244138 },139 "passed": true,140 "tolerance": {141 "rel": 1e-05142 }143 },144 {145 "description": "relative improvements = 1 - rmse_ridge/rmse_baseline (abs 1e-6)",146 "expected": null,147 "id": "rel_improvement_consistent",148 "observed": {149 "rel_improvement_vs_mean": {150 "expected": -0.031002480654837017,151 "reported": -0.031002480654837017152 },153 "rel_improvement_vs_naive": {154 "expected": 0.3649439078805412,155 "reported": 0.3649439078805412156 }157 },158 "passed": true,159 "tolerance": {160 "abs": 1e-06161 }162 },163 {164 "description": "normalisation_window == 'expanding'",165 "expected": "expanding",166 "id": "normalisation_declared",167 "observed": "expanding",168 "passed": true,169 "tolerance": null170 }171 ],172 "run_id": "RC0028",173 "schema_error": "",174 "schema_valid": true,175 "scientific_success": true,176 "task_id": "F3-conf-03"177}