FA-97856 / Dataset split isolation / Open access

Dataset split: cutoff timestamp is included in both partitions · case 01

cutoff timestamp is included in both partitions.

Verified by executionVariant 1 · 7 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

Inclusive comparison places boundary records in training.

THE FAILURE

Inclusive comparison places boundary records in training.

Unsuccessful approach: Subtracting one arbitrary timestamp unit works only for integral ticks.

Case contract

Records are [id,time]. Return sorted training IDs with time strictly less than cutoff.

Why this case matters

This toy split allocator makes membership rules explicit for reproducible evaluation fixtures.

1 / The failure

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(records, cutoff):
    return sorted(i for i,t in records if t<=cutoff)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('before',solve([['a',N-1]],N),['a'])
check('at boundary',solve([['a',N]],N),[])
check('after',solve([['a',N+1]],N),[])
check('fractional predecessor',solve([['a',N-0.5]],N),['a'])
check('unordered',solve([['b',N+2],['a',N-1]],N),['a'])
check('empty',solve([],N),[])
check('equal timestamps',solve([['a',N],['b',N-1]],N),['b'])
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
before	`['a']`	`['a']`	Passed
at boundary	`['a']`	`[]`	Failed
after	`[]`	`[]`	Passed
fractional predecessor	`['a']`	`['a']`	Passed
unordered	`['a']`	`['a']`	Passed
empty	`[]`	`[]`	Passed
equal timestamps	`['a', 'b']`	`['b']`	Failed

SHA-256 / 10c60ee51fd6819be7000c2cb3f5e982f9eee78d9f15e1052124cc8b53e2005c

2 / The unsuccessful fix

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(records, cutoff):
    return sorted(i for i,t in records if t<=cutoff-1)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('before',solve([['a',N-1]],N),['a'])
check('at boundary',solve([['a',N]],N),[])
check('after',solve([['a',N+1]],N),[])
check('fractional predecessor',solve([['a',N-0.5]],N),['a'])
check('unordered',solve([['b',N+2],['a',N-1]],N),['a'])
check('empty',solve([],N),[])
check('equal timestamps',solve([['a',N],['b',N-1]],N),['b'])
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
before	`['a']`	`['a']`	Passed
at boundary	`[]`	`[]`	Passed
after	`[]`	`[]`	Passed
fractional predecessor	`[]`	`['a']`	Failed
unordered	`['a']`	`['a']`	Passed
empty	`[]`	`[]`	Passed
equal timestamps	`['b']`	`['b']`	Passed

SHA-256 / 676224de4c9bb5fe1c4f76c2499fb4b2d9fe90910cd4a816fdc86e820afe1fc6

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 7 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Verification & scope

Inputs are small in-memory records with self-contained contracts; no claim is made about production sampling quality. 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.

Observations recorded using Python 3.12.14 at 2026-09-30T16:48:06.687926+00:00.

Case digest / 0a383e6ddef35931b8e575ae8264cce305745f23af200092885b56e2115063f4