---
title: MetricsAggregator
framework: evaluations
role: symbol
role_heading: Structure
path: evaluations/metricsaggregator
---

# MetricsAggregator

A utility for computing aggregate statistics from evaluation metrics.

## Declaration

```swift
struct MetricsAggregator
```

## Mentioned in

Designing effective evaluations Scoring with model-as-judge evaluators

## Overview

Overview let accuracy = Metric("Accuracy")

func aggregateMetrics(using aggregator: inout MetricsAggregator) {     aggregator.computeMean(of: accuracy)     aggregator.computeMaximum(of: accuracy)     aggregator.computeStandardDeviation(of: accuracy) } Use this structure to calculate summary statistics like mean, median, and standard deviation from your evaluation results. The aggregator processes metric data from a DataFrame and produces aggregated results.

## Topics

### Computing standard aggregations

- [computeMean(of:)](evaluations/metricsaggregator/computemean(of:).md)
- [computeMedian(of:)](evaluations/metricsaggregator/computemedian(of:).md)
- [computeMode(of:)](evaluations/metricsaggregator/computemode(of:).md)
- [computeMinimum(of:)](evaluations/metricsaggregator/computeminimum(of:).md)
- [computeMaximum(of:)](evaluations/metricsaggregator/computemaximum(of:).md)

### Computing variability

- [computeStandardDeviation(of:)](evaluations/metricsaggregator/computestandarddeviation(of:).md)
- [computeVariance(of:)](evaluations/metricsaggregator/computevariance(of:).md)

### Computing custom aggregations

- [custom(of:label:_:)](evaluations/metricsaggregator/custom(of:label:_:).md)

### Grouping metrics

- [group(_:_:)](evaluations/metricsaggregator/group(_:_:).md)
- [MetricsAggregator.Group](evaluations/metricsaggregator/group.md)

### Inspecting aggregate results

- [AggregateMetric](evaluations/aggregatemetric.md)
- [AggregationOperation](evaluations/aggregationoperation.md)

## See Also

### Metrics and evaluators

- [Designing specific, measurable criteria in an evaluation suite](evaluations/designing-evaluation-criteria.md)
- [Metric](evaluations/metric.md)
- [Evaluator](evaluations/evaluator.md)