Function createPrecisionEvaluator

createPrecisionEvaluator<
RecordType extends ClassificationExample = ClassificationExample,
>(
options?: PrecisionRecallFScoreOptions,
): EvaluatorBase<RecordType>
Creates a code evaluator that computes precision: of the labels the model predicted as a given class, the fraction that were actually that class.

Supports binary classification (via positiveLabel, or auto-detected when average is at its default "macro" and labels are the numeric set {0, 1}) and multi-class classification (via the average strategy).
Type Parameters
- RecordType extends ClassificationExample = ClassificationExample
Parameters
- options: PrecisionRecallFScoreOptions = {}
Returns EvaluatorBase<RecordType>
Example
```
const precision = createPrecisionEvaluator();
const result = await precision.evaluate({
  expected: ["cat", "dog", "cat", "bird"],
  output: ["cat", "cat", "cat", "bird"],
});
// { score: 5/9 }
```
- Defined in code/createPrecisionEvaluator.ts:27