modello retriever pgvector

Puoi utilizzare PostgreSQL e pgvector come implementazione del retriever. Utilizza il seguente esempio come punto di partenza e modificalo in modo che funzioni con lo schema del tuo database.

import { genkit, z, Document } from 'genkit';
import { googleAI, textEmbedding004 } from '@genkit-ai/googleai';
import { toSql } from 'pgvector';
import postgres from 'postgres';

const ai = genkit({
  plugins: [googleAI()],
});

const sql = postgres({ ssl: false, database: 'recaps' });

const QueryOptions = z.object({
  show: z.string(),
  k: z.number().optional(),
});

const sqlRetriever = ai.defineRetriever(
  {
    name: 'pgvector-myTable',
    configSchema: QueryOptions,
  },
  async (input, options) => {
    const embedding = (await ai.embed({
      embedder: textEmbedding004,
      content: input,
    }))[0].embedding;
    const results = await sql`
      SELECT episode_id, season_number, chunk as content
        FROM embeddings
        WHERE show_id = ${options.show}
        ORDER BY embedding <#> ${toSql(embedding)} LIMIT ${options.k ?? 3}
      `;
    return {
      documents: results.map((row) => {
        const { content, ...metadata } = row;
        return Document.fromText(content, metadata);
      }),
    };
  }
);

Ecco come utilizzare il retriever in un flusso:

// Simple flow to use the sqlRetriever
export const askQuestionsOnGoT = ai.defineFlow(
  {
    name: 'askQuestionsOnGoT',
    inputSchema: z.string(),
    outputSchema: z.string(),
  },
  async (inputQuestion) => {
    const docs = await ai.retrieve({
      retriever: sqlRetriever,
      query: inputQuestion,
      options: {
        show: 'Game of Thrones',
      },
    });
    console.log(docs);

    // Continue with using retrieved docs
    // in RAG prompts.
    //...
  }
);