{"abstract":"Row-normalized confusion cells use column totals.","category":"Evaluation metric aggregation","checks":6,"contract":"Return a square row-normalized confusion matrix in supplied label order; a zero-support row is all zeros.","contract_signature":"truth, pred, labels","evaluation_group":"w3-eval-confusion-row-normalization","failed_approach":"Adding one to every row denominator obscures perfect rows with small support.","family":"w3-evaluation-metric-aggregation-confusion-row-normalization","id":"FA-100026","implementations":{"attempt":{"sha256":"4e7014a88bf63365699402c9a4ba65ac9a063805fc9175341b9c54b438ef087e","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(truth, pred, labels):\n    return [[sum(a==i and b==j for a,b in zip(truth,pred))/(sum(a==i for a in truth)+1) for j in labels] for i in labels]\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('row distributions',solve([0,0,1],[0,1,1],[0,1]),[[.5,.5],[0,1]])\ncheck('perfect diagonal',solve([0,1],[0,1],[0,1]),[[1,0],[0,1]])\ncheck('absent row',solve([0],[0],[0,1]),[[1,0],[0,0]])\ncheck('empty labels',solve([0],[0],[]),[])\ncheck('empty data',solve([],[],[0,1]),[[0,0],[0,0]])\ncheck('reversed label order',solve([N,N+1],[N+1,N],[N+1,N]),[[0,1],[1,0]])\nprint(json.dumps({\"observations\": observations, \"passed\": all(x[\"passed\"] for x in observations)}, ensure_ascii=False))\nraise SystemExit(0 if all(x[\"passed\"] for x in observations) else 1)\n"},"broken":{"sha256":"4738d6eac39c39010291ba1fe7e85478c61d54f82713b4f1d33165227aa11c36","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(truth, pred, labels):\n    return [[sum(a==i and b==j for a,b in zip(truth,pred))/(sum(b==i for b in pred) or 1) for j in labels] for i in labels]\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('row distributions',solve([0,0,1],[0,1,1],[0,1]),[[.5,.5],[0,1]])\ncheck('perfect diagonal',solve([0,1],[0,1],[0,1]),[[1,0],[0,1]])\ncheck('absent row',solve([0],[0],[0,1]),[[1,0],[0,0]])\ncheck('empty labels',solve([0],[0],[]),[])\ncheck('empty data',solve([],[],[0,1]),[[0,0],[0,0]])\ncheck('reversed label order',solve([N,N+1],[N+1,N],[N+1,N]),[[0,1],[1,0]])\nprint(json.dumps({\"observations\": observations, \"passed\": all(x[\"passed\"] for x in observations)}, ensure_ascii=False))\nraise SystemExit(0 if all(x[\"passed\"] for x in observations) else 1)\n"}},"limitations":"Inputs are finite toy labels and scores; this defines no benchmark protocol, production metric library, or model quality claim. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.","method":"Deterministic executable model with adversarial boundary fixtures.","provenance":{"created_by":"Failure Map","dependencies":"Python standard library","family":"w3-evaluation-metric-aggregation-confusion-row-normalization","generated_at":"2026-09-30T16:48:30.810592+00:00","license":"CC0-1.0","python":"3.12.14","seed":1,"split":"open-access"},"relevance":"A small explicit evaluation contract makes aggregation behavior reproducible across labeled examples.","root_cause":"Each actual class row is divided by predicted support for its class.","sha256":"779a7fd07bda86dd8fa8a5dd290a5c8ff88ba56c76887ec8d83f99bb08abbe66","title":"Row-normalized confusion cells use column totals · case 01","variant":1,"variant_policy":"The fixture code references N, supplied as 1 through 5. Fixed boundary checks may repeat. These records share one mechanism and must remain together in evaluation splits.","verified":true,"visibility":"public","verification":{"attempt":{"elapsed_ms":30.218,"exit_code":1,"observations":[{"actual":[[0.3333333333333333,0.3333333333333333],[0.0,0.5]],"check":"row distributions","expected":[[0.5,0.5],[0,1]],"passed":false},{"actual":[[0.5,0.0],[0.0,0.5]],"check":"perfect diagonal","expected":[[1,0],[0,1]],"passed":false},{"actual":[[0.5,0.0],[0.0,0.0]],"check":"absent row","expected":[[1,0],[0,0]],"passed":false},{"actual":[],"check":"empty labels","expected":[],"passed":true},{"actual":[[0.0,0.0],[0.0,0.0]],"check":"empty data","expected":[[0,0],[0,0]],"passed":true},{"actual":[[0.0,0.5],[0.5,0.0]],"check":"reversed label order","expected":[[0,1],[1,0]],"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"row distributions\", \"actual\": [[0.3333333333333333, 0.3333333333333333], [0.0, 0.5]], \"expected\": [[0.5, 0.5], [0, 1]], \"passed\": false}, {\"check\": \"perfect diagonal\", \"actual\": [[0.5, 0.0], [0.0, 0.5]], \"expected\": [[1, 0], [0, 1]], \"passed\": false}, {\"check\": \"absent row\", \"actual\": [[0.5, 0.0], [0.0, 0.0]], \"expected\": [[1, 0], [0, 0]], \"passed\": false}, {\"check\": \"empty labels\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"empty data\", \"actual\": [[0.0, 0.0], [0.0, 0.0]], \"expected\": [[0, 0], [0, 0]], \"passed\": true}, {\"check\": \"reversed label order\", \"actual\": [[0.0, 0.5], [0.5, 0.0]], \"expected\": [[0, 1], [1, 0]], \"passed\": false}], \"passed\": false}\n"},"broken":{"elapsed_ms":26.618,"exit_code":1,"observations":[{"actual":[[1.0,1.0],[0.0,0.5]],"check":"row distributions","expected":[[0.5,0.5],[0,1]],"passed":false},{"actual":[[1.0,0.0],[0.0,1.0]],"check":"perfect diagonal","expected":[[1,0],[0,1]],"passed":true},{"actual":[[1.0,0.0],[0.0,0.0]],"check":"absent row","expected":[[1,0],[0,0]],"passed":true},{"actual":[],"check":"empty labels","expected":[],"passed":true},{"actual":[[0.0,0.0],[0.0,0.0]],"check":"empty data","expected":[[0,0],[0,0]],"passed":true},{"actual":[[0.0,1.0],[1.0,0.0]],"check":"reversed label order","expected":[[0,1],[1,0]],"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"row distributions\", \"actual\": [[1.0, 1.0], [0.0, 0.5]], \"expected\": [[0.5, 0.5], [0, 1]], \"passed\": false}, {\"check\": \"perfect diagonal\", \"actual\": [[1.0, 0.0], [0.0, 1.0]], \"expected\": [[1, 0], [0, 1]], \"passed\": true}, {\"check\": \"absent row\", \"actual\": [[1.0, 0.0], [0.0, 0.0]], \"expected\": [[1, 0], [0, 0]], \"passed\": true}, {\"check\": \"empty labels\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"empty data\", \"actual\": [[0.0, 0.0], [0.0, 0.0]], \"expected\": [[0, 0], [0, 0]], \"passed\": true}, {\"check\": \"reversed label order\", \"actual\": [[0.0, 1.0], [1.0, 0.0]], \"expected\": [[0, 1], [1, 0]], \"passed\": true}], \"passed\": false}\n"}},"member_only":{"stages":["fixed"],"fields":["implementations.fixed","verification.fixed","harness","repair"],"note":"The verified repair, its recorded checks, the repair description, and the scoring harness are available to members."}}