Description
Choose the range or absolute arithmetic mean of nonempty referenceValues as scale.
Normalizing a metric divides by a declared reference scale or changes it according to its dimensional exponent. Different scales can produce different comparisons from identical errors.
Choose the range or absolute arithmetic mean of nonempty referenceValues as scale.
When to use this tool
Use Metrics: Nrmse Reference Scale Choice when you need a transparent reference-scale calculation.
How to use the tool and how it works
- Enter rmse, reference, referenceValues. 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.
Choose the range or absolute arithmetic mean of nonempty referenceValues as scale. Require positive scale, then divide nonnegative rmse by it.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| rmse | Rmse | Required. Finite number. |
| reference | Reference | Required. Text; one of "range", "absolute-mean". Implementation limit: 8192 Unicode code points and 16384 UTF-16 code units. |
| referenceValues | Reference Values | Required. Array of finite number. Implementation limit: 128 items. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| result | Result | Required. Object. |
| result.scale | Scale | Calculated result member: number. |
| result.normalizedRmse | Normalized Rmse | Calculated result member: number. |
{
"input": {
"rmse": 2,
"reference": "range",
"referenceValues": [
2,
6
]
},
"output": {
"result": {
"scale": 4,
"normalizedRmse": 0.5
}
}
}Limitations and assumptions
- Normalization depends on the chosen reference set; constant ranges and zero absolute means are rejected, and RMSE itself is supplied.
- 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
Report the chosen scale and reference set alongside the original metric, especially when scales are near zero.
References
Similar or alternative tools
- Scaled Error Seasonal Denominator Zero
Divide nonnegative absoluteError by seasonalScale when positive.
- Error Metric Unit Scaling Exponent
Multiply nonnegative metric by positive unitScale raised to exponent.