---
title: "evaluation(on:)"
framework: createml
role: symbol
role_heading: Instance Method
path: "createml/mlregressor/evaluation(on:)-7pirm"
---

# evaluation(on:)

Evaluates the classifier on the provided labeled data.

## Declaration

```swift
func evaluation(on labeledData: MLDataTable) -> MLRegressorMetrics
```

## Parameters

- `labeledData`: An MLDataTable to evaluate the trained model on.

## Return Value

Return Value Metrics that describe the maximum error (maximumError) or the average error (rootMeanSquaredError).

## Discussion

Discussion Evaluation should be done on a testing data set that the model has not seen as part of the training or validation data sets. The data should have feature columns with identical name and type to the training data, as well as a labels column with the same name.