Class: Evaluation<R, E, M>

TypeScript SDK 레퍼런스

weave / Evaluation

scorers 세트와 Datasets 를 포함하는 evaluation을 설정합니다.

evaluation.evaluate(model)을 호출하면 Datasets 의 각 행이 모델로 전달되며, 이때 Datasets 의 컬럼 이름과 model.predict의 인수(argument) 이름을 매칭합니다.

그 후 모든 scorers를 호출하고 결과를 Weave 에 저장합니다.

예시

// 예시 데이터들을 모아 Datasets 생성
const dataset = new weave.Dataset({
  id: 'my-dataset',
  rows: [
    { question: 'What is the capital of France?', expected: 'Paris' },
    { question: 'Who wrote "To Kill a Mockingbird"?', expected: 'Harper Lee' },
    { question: 'What is the square root of 64?', expected: '8' },
  ],
});

// 커스텀 scoring 함수 정의
const scoringFunction = weave.op(function isEqual({ modelOutput, datasetRow }) {
  return modelOutput == datasetRow.expected;
});

// 평가할 함수 정의
const model = weave.op(async function alwaysParisModel({ question }) {
  return 'Paris';
});

// 평가 시작
const evaluation = new weave.Evaluation({
  id: 'my-evaluation',
  dataset: dataset,
  scorers: [scoringFunction],
});

const results = await evaluation.evaluate({ model });

​

Type parameters

Name	Type
R	extends DatasetRow
E	extends DatasetRow
M	M

​

Hierarchy

• WeaveObject

  ↳ Evaluation

​

목차

​

생성자 (Constructors)

• constructor

​

속성 (Properties)

• __savedRef

​

Accessors

• description
• name

​

메소드 (Methods)

• evaluate
• predictAndScore
• saveAttrs

​

생성자 (Constructors)

​

constructor

• new Evaluation<R, E, M>(parameters): Evaluation<R, E, M>

​

Type parameters

Name	Type
R	extends DatasetRow
E	extends DatasetRow
M	M

​

파라미터 (Parameters)

Name	Type
parameters	EvaluationParameters<R, E, M>

​

반환값 (Returns)

Evaluation<R, E, M>

​

Overrides

WeaveObject.constructor

​

Defined in

evaluation.ts:148

​

속성 (Properties)

​

__savedRef

• Optional __savedRef: ObjectRef | Promise<ObjectRef>

​

Inherited from

WeaveObject.__savedRef

​

Defined in

weaveObject.ts:73

​

Accessors

​

description

• get description(): undefined | string

​

반환값 (Returns)

undefined | string

​

Inherited from

WeaveObject.description

​

Defined in

weaveObject.ts:100

---

​

name

• get name(): string

​

반환값 (Returns)

string

​

Inherited from

WeaveObject.name

​

Defined in

weaveObject.ts:96

​

메소드 (Methods)

​

evaluate

▸ evaluate(«destructured»): Promise<Record<string, any>>

​

파라미터 (Parameters)

Name	Type	Default value
«destructured»	Object	undefined
› maxConcurrency?	number	5
› model	WeaveCallable<(…args: [{ datasetRow: R }]) => Promise<M>>	undefined
› nTrials?	number	1

​

반환값 (Returns)

Promise<Record<string, any>>

​

Defined in

evaluation.ts:163

---

​

predictAndScore

▸ predictAndScore(«destructured»): Promise<{ model_latency: number = modelLatency; model_output: any = modelOutput; model_success: boolean = !modelError; scores: { [key: string]: any; } }>

​

파라미터 (Parameters)

Name	Type
«destructured»	Object
› columnMapping?	ColumnMapping<R, E>
› example	R
› model	WeaveCallable<(…args: [{ datasetRow: E }]) => Promise<M>>

​

반환값 (Returns)

Promise<{ model_latency: number = modelLatency; model_output: any = modelOutput; model_success: boolean = !modelError; scores: { [key: string]: any; } }>

​

Defined in

evaluation.ts:231

---

​

saveAttrs

▸ saveAttrs(): Object

​

반환값 (Returns)

Object

​

Inherited from

WeaveObject.saveAttrs

​

Defined in

weaveObject.ts:77