如何使用 LangChain 和 Redis 构建一个 GenAI 聊天机器人

import { Document } from 'langchain/document';
import { OpenAIEmbeddings } from 'langchain/embeddings/openai';
import { RedisVectorStore } from 'langchain/vectorstores/redis';

/**
 * Adds OpenAI embeddings to Redis for the given products.
 *
 * @param _products - An array of (ecommerce) products.
 * @param _redisClient - The Redis client used to connect to the Redis server.
 * @param _openAIApiKey - The API key for accessing the OpenAI service.
 */
const addOpenAIEmbeddingsToRedis = async (
  _products,
  _redisClient,
  _openAIApiKey,
) => {
  if (_products?.length > 0 && _redisClient && _openAIApiKey) {
    // Check if the data is already seeded
    const existingKeys = await _redisClient.keys('openAIProducts:*');
    if (existingKeys.length > 0) {
      console.log('seeding openAIEmbeddings skipped !');
      return;
    }

    const vectorDocs: Document[] = [];
    // Create a document for each product
    for (let product of _products) {
      let doc = new Document({
        metadata: {
          productId: product.productId,
        },
        pageContent: ` Product details are as follows:
                productId: ${product.productId}.
    
                productDisplayName: ${product.productDisplayName}.
                
                price: ${product.price}.
    
                variantName: ${product.variantName}.
    
                brandName: ${product.brandName}.
    
                ageGroup: ${product.ageGroup}.
    
                gender: ${product.gender}.
    
                productColors: ${product.productColors}
    
                Category:  ${product.displayCategories}, ${product.masterCategory_typeName} - ${product.subCategory_typeName}
    
                productDescription:  ${product.productDescriptors_description_value}`,
      });

      vectorDocs.push(doc);
    }

    // Create a new OpenAIEmbeddings instance
    const embeddings = new OpenAIEmbeddings({
      openAIApiKey: _openAIApiKey,
    });
    // Add the documents to the RedisVectorStore
    const vectorStore = await RedisVectorStore.fromDocuments(
      vectorDocs,
      embeddings,
      {
        redisClient: _redisClient,
        indexName: 'openAIProductsIdx',
        keyPrefix: 'openAIProducts:',
      },
    );
    console.log('seeding OpenAIEmbeddings completed');
  }
};