Description
Compare one confusion row's diagonal with all off-diagonal mass.
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.
Compare one confusion row's diagonal with all off-diagonal mass.
When to use this tool
Use Confusion Row Diagonal Dominance Gap when checking the stated confusion-count, metric contribution or abstention-denominator convention.
How to use the tool and how it works
- Enter row, diagonalIndex. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read gap, dominates. Inspect dominates before interpreting the associated values.
Compare one confusion row's diagonal with all off-diagonal mass. Require nonempty nonnegative row and valid diagonalIndex; gap = diagonal-sum(other entries), dominates means gap >= 0.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| row | Row | Required. Array of safe integer; maximum items 256. |
| diagonalIndex | Diagonal Index | Required. Safe integer; minimum 0. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| gap | Signed difference under the stated comparison convention, with matching units. | Required. Safe integer. |
| dominates | Dominates | Required. Boolean. |
{
"input": {
"row": [
8,
2,
1
],
"diagonalIndex": 0
},
"output": {
"gap": 5,
"dominates": true
}
}Limitations and assumptions
- Equality counts as dominance; this does not validate a confusion matrix or prove class/model quality.
- 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
Similar or alternative tools
- Confusion Cell Expected Independence Term
Return one expected confusion count under independence.
- Binary Confusion Transpose Role Change
Transpose the roles of binary confusion axes.