Classification: Abstention Correct Denominator Tool

Description

Return accuracy under an explicit abstention denominator.

Classifier metrics count particular outcomes on a declared population and positive-label convention. Precision, recall and accepted-only measures use different denominators. A reporting value at an empty denominator is a policy choice, while one cell, derivative or contribution does not constitute a complete performance evaluation.

Return accuracy under an explicit abstention denominator.

When to use this tool

Use Classification: Abstention Correct Denominator when checking the stated confusion-count, metric contribution or abstention-denominator convention.

How to use the tool and how it works

Enter correct, attempted, abstained, denominatorPolicy. Use the units and conventions below.
Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
Read result. Expand result to inspect the returned fields.

Return accuracy under an explicit abstention denominator. Require bounded whole correct <= attempted and abstained >= 0; denominator adds abstentions only for include-abstentions, defined tests positive denominator, accuracy divides or returns zero.^[1]

Classifier metrics use different outcome supports and denominator conventions. A zero-denominator reporting policy does not supply missing observed outcomes.^[2]

Input field	Meaning and units	Accepted input
correct	Correct	Required. Safe integer. Calculation count range: 0-1000000 inclusive.
attempted	Attempted	Required. Safe integer. Calculation count range: 0-1000000 inclusive.
abstained	Abstained	Required. Safe integer. Calculation count range: 0-1000000 inclusive.
denominatorPolicy	Denominator Policy	Required. Text; one of "include-abstentions", "exclude-abstentions". Implementation limit: 8192 Unicode code points and 16384 UTF-16 code units.

Output field	Interpretation	Returned shape
result	Returned calculated value or nested object under the stated output convention.	Required. Object.
result.denominator	Denominator	Calculated result member: number.
result.defined	Reporting flag under the explicit denominator or information policy above; inspect any zero placeholder or policy-defined value.	Calculated result member: boolean.
result.accuracy	Accuracy	Calculated result member: number.

json

{
  "input": {
    "correct": 6,
    "attempted": 8,
    "abstained": 2,
    "denominatorPolicy": "include-abstentions"
  },
  "output": {
    "result": {
      "denominator": 10,
      "defined": true,
      "accuracy": 0.6
    }
  }
}

Limitations and assumptions

The policy changes the reported metric; excluded abstentions and zero-denominator placeholders must not be mistaken for observed accuracy.
Outputs are nested inside result. 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

Evaluate the complete confusion table, relevant class supports and abstention coverage together; use suitable uncertainty and validation methods for conclusions about predictive performance.

References

Calculation and input contract: classification/abstention-correct-denominator — tool.io
Model evaluation: classification metrics — scikit-learn developers

Similar or alternative tools

Abstention Zero Coverage Policy
Report coverage and accepted accuracy under a zero-coverage policy.
Binary Abstention Coverage Count
Return acceptance coverage from positive, negative and abstained counts.