{"abstract":"Top k accuracy divides by candidate slots.","category":"Evaluation metric aggregation","checks":6,"contract":"Given target labels and candidate rows, return the fraction of rows containing their target; empty inputs yield 0.","contract_signature":"targets, candidates","evaluation_group":"w3-eval-topk-accuracy","failed_approach":"Counting duplicate candidate slots can credit a single example more than once.","family":"w3-evaluation-metric-aggregation-topk-accuracy","id":"FA-99991","implementations":{"attempt":{"sha256":"ff3f1704fba029b06ca8fa7aa1d80a5c5ec702cc97f1683fb9b98ea6622de8aa","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(targets, candidates):\n    hits=sum(t in row for t,row in zip(targets,candidates)); return hits/(len(targets)+1) if targets else 0\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('ragged candidates',solve([1,2],[[1,0,3],[2]]),1)\ncheck('one miss',solve([1,2],[[1],[0]]),.5)\ncheck('duplicate hit counted once',solve([1],[[1,1,1]]),1)\ncheck('empty',solve([],[]),0)\ncheck('singleton hit',solve([N],[[N]]),1)\ncheck('singleton miss',solve([N],[[N+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":"b70ffac04f76f8cd5581e333d829c11fe5768f6251a85ede9a65c1eade690387","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(targets, candidates):\n    hits=sum(t in row for t,row in zip(targets,candidates)); slots=sum(len(row) for row in candidates); return hits/slots if slots else 0\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('ragged candidates',solve([1,2],[[1,0,3],[2]]),1)\ncheck('one miss',solve([1,2],[[1],[0]]),.5)\ncheck('duplicate hit counted once',solve([1],[[1,1,1]]),1)\ncheck('empty',solve([],[]),0)\ncheck('singleton hit',solve([N],[[N]]),1)\ncheck('singleton miss',solve([N],[[N+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-topk-accuracy","generated_at":"2026-09-30T16:48:30.355904+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":"The hit rate is reduced by padded candidate positions.","sha256":"c463398b020552dff6c14d506df2931021b56660f097251964a04d17a325f534","title":"Top k accuracy divides by candidate slots · 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":30.216,"exit_code":1,"observations":[{"actual":0.6666666666666666,"check":"ragged candidates","expected":1,"passed":false},{"actual":0.3333333333333333,"check":"one miss","expected":0.5,"passed":false},{"actual":0.5,"check":"duplicate hit counted once","expected":1,"passed":false},{"actual":0,"check":"empty","expected":0,"passed":true},{"actual":0.5,"check":"singleton hit","expected":1,"passed":false},{"actual":0.0,"check":"singleton miss","expected":0,"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"ragged candidates\", \"actual\": 0.6666666666666666, \"expected\": 1, \"passed\": false}, {\"check\": \"one miss\", \"actual\": 0.3333333333333333, \"expected\": 0.5, \"passed\": false}, {\"check\": \"duplicate hit counted once\", \"actual\": 0.5, \"expected\": 1, \"passed\": false}, {\"check\": \"empty\", \"actual\": 0, \"expected\": 0, \"passed\": true}, {\"check\": \"singleton hit\", \"actual\": 0.5, \"expected\": 1, \"passed\": false}, {\"check\": \"singleton miss\", \"actual\": 0.0, \"expected\": 0, \"passed\": true}], \"passed\": false}\n"},"broken":{"elapsed_ms":27.377,"exit_code":1,"observations":[{"actual":0.5,"check":"ragged candidates","expected":1,"passed":false},{"actual":0.5,"check":"one miss","expected":0.5,"passed":true},{"actual":0.3333333333333333,"check":"duplicate hit counted once","expected":1,"passed":false},{"actual":0,"check":"empty","expected":0,"passed":true},{"actual":1.0,"check":"singleton hit","expected":1,"passed":true},{"actual":0.0,"check":"singleton miss","expected":0,"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"ragged candidates\", \"actual\": 0.5, \"expected\": 1, \"passed\": false}, {\"check\": \"one miss\", \"actual\": 0.5, \"expected\": 0.5, \"passed\": true}, {\"check\": \"duplicate hit counted once\", \"actual\": 0.3333333333333333, \"expected\": 1, \"passed\": false}, {\"check\": \"empty\", \"actual\": 0, \"expected\": 0, \"passed\": true}, {\"check\": \"singleton hit\", \"actual\": 1.0, \"expected\": 1, \"passed\": true}, {\"check\": \"singleton miss\", \"actual\": 0.0, \"expected\": 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."}}