FA-99166 / Checkpoint tensor mapping / Open access

Checkpoint mapping: renaming collisions are not rejected · case 01

renaming collisions are not rejected.

Verified by executionVariant 1 · 6 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

No uniqueness check follows exact renaming, so two source tensors can map to one destination name.

THE FAILURE

No uniqueness check follows exact renaming, so two source tensors can map to one destination name.

Unsuccessful approach: Dictionary conversion retains only the last colliding value, losing the evidence needed to reject the conflict.

Case contract

Apply exact-name renames to [name,value] pairs and return sorted pairs; if two inputs map to one output name, return [].

Why this case matters

A bounded offline model of checkpoint tensor bookkeeping in a training workflow.

1 / The failure

Exit 1

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

N = 1
observations = []
def solve(pairs, renames):
    return sorted((renames.get(k,k),v) for k,v in pairs)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('collision', solve(*([['a', 3], ['b', 2]], {'a': 'x', 'b': 'x'})), [])
check('simple', solve(*([['a', 3]], {'a': 'z'})), [['z', 3]])
check('no map', solve(*([['a', 3]], {})), [['a', 3]])
check('empty', solve(*([], {'a': 'x'})), [])
check('identity', solve(*([['a', 1], ['b', 2]], {})), [['a', 1], ['b', 2]])
check('one collision', solve(*([['a', 1], ['b', 2], ['c', 3]], {'a': 'c', 'b': '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
collision	`[['x', 2], ['x', 3]]`	`[]`	Failed
simple	`[['z', 3]]`	`[['z', 3]]`	Failed
no map	`[['a', 3]]`	`[['a', 3]]`	Failed
empty	`[]`	`[]`	Passed
identity	`[['a', 1], ['b', 2]]`	`[['a', 1], ['b', 2]]`	Failed
one collision	`[['b', 2], ['c', 1], ['c', 3]]`	`[]`	Failed

SHA-256 / 37ad675d614ea7019b2aacf51c2170e67bc95206d8b7b19d8bb3d75bd963066e

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(pairs, renames):
    return sorted({renames.get(k,k):v for k,v in pairs}.items())
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('collision', solve(*([['a', 3], ['b', 2]], {'a': 'x', 'b': 'x'})), [])
check('simple', solve(*([['a', 3]], {'a': 'z'})), [['z', 3]])
check('no map', solve(*([['a', 3]], {})), [['a', 3]])
check('empty', solve(*([], {'a': 'x'})), [])
check('identity', solve(*([['a', 1], ['b', 2]], {})), [['a', 1], ['b', 2]])
check('one collision', solve(*([['a', 1], ['b', 2], ['c', 3]], {'a': 'c', 'b': '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
collision	`[['x', 2]]`	`[]`	Failed
simple	`[['z', 3]]`	`[['z', 3]]`	Failed
no map	`[['a', 3]]`	`[['a', 3]]`	Failed
empty	`[]`	`[]`	Passed
identity	`[['a', 1], ['b', 2]]`	`[['a', 1], ['b', 2]]`	Failed
one collision	`[['b', 2], ['c', 3]]`	`[]`	Failed

SHA-256 / 3f2274f4e6a9854bd9082df870cde9908c7f6eff830d03ec5a2da09ecaa2c470

HELD IN THE MEMBER ARCHIVE

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

This mechanism has 6 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 already decoded Python values; this toy contract makes no claim about a framework, file format, or production checkpoint. 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 2026-09-30T16:48:21.244522+00:00.

Case digest / 0f9af26c8d4e0292c25891a22e8b00dfa8f02596d1ecd4f71599d7d4d2c18d8e