public static fromMean(meanLoss as Double, samples as ULong, correct as ULong = 0ul, hasAccuracy as Boolean = false) as Evaluation
Builds metrics from an already reduced mean without multiplying it by the sample count, which could overflow even when the mean is finite.
| Parameter | Description |
|---|---|
meanLoss | Finite non-negative mean objective. |
samples | Positive observation count. |
correct | Correct predictions, used only when hasAccuracy is true. |
hasAccuracy | False for regression; true for classification. |
Independent evaluation.
result = Evaluation.fromMean(0.25, 100ul)