Description
Return signed and absolute ordinal class-index errors.
Ordinal labels have an order, but adjacent categories need not be equally spaced measurements. Index distances, cumulative targets and weighted medians preserve a specified ordering while probability normalization, equality at a median boundary and distance costs remain distinct conventions.
Return signed and absolute ordinal class-index errors.
When to use this tool
Use Classification: Ordinal Step Error Distance when evaluating the stated ordered-label target, distance or mass convention.
How to use the tool and how it works
- Enter actualIndex, predictedIndex, classCount. 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 signed and absolute ordinal class-index errors. Require bounded positive whole classCount and actual/predicted indexes in [0,classCount); signedSteps = predicted-actual, absoluteSteps is its magnitude.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| actualIndex | Actual Index | Required. Safe integer. Calculation range: 0 through classCount - 1, inclusive. |
| predictedIndex | Predicted Index | Required. Safe integer. Calculation range: 0 through classCount - 1, inclusive. |
| classCount | Class Count | Required. Safe integer. Calculation range: 1 through 1000000, inclusive. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| result | Returned calculated value or nested object under the stated output convention. | Required. Object. |
| result.signedSteps | Signed Steps | Calculated result member: number. |
| result.absoluteSteps | Absolute Steps | Calculated result member: number. |
{
"input": {
"actualIndex": 1,
"predictedIndex": 3,
"classCount": 4
},
"output": {
"result": {
"signedSteps": 2,
"absoluteSteps": 2
}
}
}Limitations and assumptions
- Each adjacent index has one unit cost; ordinal labels need not be equal-interval measurements.
- 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
Use meaningful ordered labels and an appropriate ordinal model; evaluate prediction quality with costs and probability assumptions justified for the actual categories.
References
Similar or alternative tools
- Ordinal Probability Mass Interval
Return probability mass over a closed ordinal-index interval.
- Ordinal Median Class Boundary
Select a weighted ordinal median under an equality policy.