Description
Subtract log1p(actual) from log1p(forecast), then return that residual, its square, and its absolute value.
Logarithmic errors compare multiplicative differences. Positive-domain requirements and any added offset change both valid inputs and the interpretation of small values.
Subtract log1p(actual) from log1p(forecast), then return that residual, its square, and its absolute value.
When to use this tool
Use Metrics: Rmsle Single Pair Domain Check when you need a single log-error term or a logarithm-domain check.
How to use the tool and how it works
- Enter actual, forecast. 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.
Require nonnegative actual and forecast. Subtract log1p(actual) from log1p(forecast), then return that residual, its square, and its absolute value.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| actual | Observed value on the same scale and unit as its prediction. | Required. Finite number. |
| forecast | Forecast value on the same scale and unit as actual. | Required. Finite number. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| result | Result | Required. Object. |
| result.logResidual | Log Residual | Calculated result member: number. |
| result.squaredLogResidual | Squared Log Residual | Calculated result member: number. |
| result.rootSingleSquaredLogError | Root Single Squared Log Error | Calculated result member: number. |
{
"input": {
"actual": 0,
"forecast": 1
},
"output": {
"result": {
"logResidual": 0.6931471805599453,
"squaredLogResidual": 0.4804530139182014,
"rootSingleSquaredLogError": 0.6931471805599453
}
}
}Limitations and assumptions
- This is one pair's log error, not an aggregate RMSLE; the unit-dependent +1 convention is fixed.
- 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
Choose a documented offset and aggregation policy and compare errors on the original scale where useful.
References
Similar or alternative tools
- Rmsle Log Offset Domain Check
Add positive offset separately to actual and predicted.
- Absolute Log Error Single Pair
Require positive actual and predicted and return abs(log(predicted / actual)).