Verification Score Source Gap Tool

Description

Compare score and model-version records.

Biometric research records separate subject IDs, template dimensions, image geometry, enrollment support and attempt budgets. Local counts and comparisons can reveal inconsistencies without reading biometric templates, detecting a physical trait or authenticating a person’s identity.

Compare score and model-version records.

When to use this tool

Use Verification Score Source Gap when checking the stated biometric record, count, support, geometry or budget convention.

How to use the tool and how it works

Enter scores, sourceScores, modelVersions, requiredVersion. Follow any array-alignment requirements below. Use the units and conventions below.
Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
Read scoreResiduals, versionMismatches, maximumAbsoluteResidual. Use the output convention below.

Compare score and model-version records. Require aligned scores, sourceScores and modelVersions; residuals = scores-sourceScores, count versions differing from requiredVersion and return maximum absolute residual, zero for empty arrays.^[1]

Input field	Meaning and units	Accepted input
scores	Scores	Required. Array of finite number; maximum items 256.
sourceScores	Source Scores	Required. Array of finite number; maximum items 256.
modelVersions	Model Versions	Required. Array of text; maximum UTF-16 length 8192; maximum items 256.
requiredVersion	Required Version	Required. Text; maximum UTF-16 length 8192.

Output field	Interpretation	Returned shape
scoreResiduals	Score Residuals	Required. Array of finite number; maximum items 256.
versionMismatches	Version Mismatches	Required. Safe integer.
maximumAbsoluteResidual	Maximum Absolute Residual	Required. Finite number.

json

{
  "input": {
    "scores": [
      2,
      4
    ],
    "sourceScores": [
      1,
      4
    ],
    "modelVersions": [
      "v2",
      "v1"
    ],
    "requiredVersion": "v2"
  },
  "output": {
    "scoreResiduals": [
      1,
      0
    ],
    "versionMismatches": 1,
    "maximumAbsoluteResidual": 1
  }
}

Limitations and assumptions

Matching records do not authenticate an identity, validate a model or prove an unbiased evaluation.
Inputs must satisfy both the field types and the cross-field conditions above. Invalid input causes an error; a successful false flag or documented placeholder is a different result.
Numeric calculations use binary floating point unless exact decimal-integer arithmetic is explicitly stated. No hidden measurement tolerance or decimal rounding is added; nonfinite or out-of-contract results are rejected. Numeric signed-zero results are exported as zero.

Alternative or Complementary methods

Use verified acquisition/enrollment records and the complete appropriate format and evaluation process for actual identity, quality, capacity or privacy decisions.

References

Calculation and input contract: biometrics/verification-score-source-gap — tool.io

Similar or alternative tools

Biometric Research Subject Split Overlap
Return distinct subject-ID overlaps among research splits.
Biometric Presentation Record Caller Code
Check a presentation code against a supplied vocabulary.