rerankWithScorer()

rerankWithScorer() 函数通过结合语义相关性、向量相似性和基于位置的评分，为向量搜索结果提供了高级的重新排序功能。

🌐 The rerankWithScorer() function provides advanced reranking capabilities for vector search results by combining semantic relevance, vector similarity, and position-based scoring.

function rerankWithScorer({
  results: QueryResult[],
  query: string,
  scorer: RelevanceScoreProvider,
  options?: RerankerFunctionOptions,
}): Promise<RerankResult[]>;

使用示例

🌐 Usage Example

import { rerankWithScorer as rerank, CohereRelevanceScorer } from "@mastra/rag";

const scorer = new CohereRelevanceScorer("rerank-v3.5");

const rerankedResults = await rerank({
  results: vectorSearchResults,
  query: "How do I deploy to production?",
  scorer,
  options: {
    weights: {
      semantic: 0.5,
      vector: 0.3,
      position: 0.2,
    },
    topK: 3,
  },
});

参数

🌐 Parameters

results:

QueryResult[]

The vector search results to rerank

query:

string

The search query text used to evaluate relevance

scorer:

RelevanceScoreProvider

The relevance scorer to use for reranking

options?:

RerankerFunctionOptions

Options for the reranking model

rerankWithScorer 函数接受来自 @mastra/rag 的任意 RelevanceScoreProvider。

🌐 The rerankWithScorer function accepts any RelevanceScoreProvider from @mastra/rag.

注意: 为了使语义评分在重新排序时正常工作，每个结果必须在其 metadata.text 字段中包含文本内容。

RerankerFunctionOptions

weights?:

WeightConfig

Weights for different scoring components (must add up to 1)

number

semantic?:

number (default: 0.4)

Weight for semantic relevance

number

vector?:

number (default: 0.4)

Weight for vector similarity

number

position?:

number (default: 0.2)

Weight for position-based scoring

queryEmbedding?:

number[]

Embedding of the query

topK?:

number

= 3

Number of top results to return

返回

🌐 Returns

该函数返回一个 RerankResult 对象数组：

🌐 The function returns an array of RerankResult objects:

result:

QueryResult

The original query result

score:

number

Combined reranking score (0-1)

details:

ScoringDetails

Detailed scoring information

ScoringDetails

semantic:

number

Semantic relevance score (0-1)

vector:

number

Vector similarity score (0-1)

position:

number

Position-based score (0-1)

queryAnalysis?:

object

Query analysis details

number

magnitude:

Magnitude of the query

number[]

dominantFeatures:

Dominant features of the query

相关

🌐 Related

• 创建向量查询工具