.NET + AI | Semantic Kernel 向量搜索使用指南

#r "nuget:Microsoft.SemanticKernel.Connectors.InMemory,*-*"
#r "nuget:Microsoft.Extensions.AI,*-*"
#r "nuget:Microsoft.Extensions.AI.OpenAI,*-*"
#r "nuget:Azure.AI.OpenAI,*-*"

using System.ClientModel;
using Microsoft.Extensions.AI;
using Azure.AI.OpenAI;

var endpoint = "YOUR_AZURE_OPENAI_ENDPOINT";
var apikey = "YOUR_API_KEY";

var azureClinet = new AzureOpenAIClient(
    endpoint: new Uri(endpoint),
    credential: new ApiKeyCredential("apikey")
);

var generator = azureClinet.GetEmbeddingClient("text-embedding-3-small")
                            .AsIEmbeddingGenerator();

var embeddings = await generator.GenerateAsync("What is AI?");

embeddings.Display();

using Microsoft.Extensions.VectorData;

// The data model
internalclassFinanceInfo
{
    [VectorStoreRecordKey]
    publicstring Key { get; set; } = string.Empty;

    [VectorStoreRecordData]
    publicstring Text { get; set; } = string.Empty;

    // Note that the vector property is typed as a string, and
    // its value is derived from the Text property. The string
    // value will however be converted to a vector on upsert and
    // stored in the database as a vector.
    [VectorStoreRecordVector(1536)]
    publicstring Embedding => this.Text;
}

using Microsoft.SemanticKernel.Connectors.InMemory;

// Use the embedding generator with the vector store.

// use in memory vector store
var vectorStore = new InMemoryVectorStore(new() { EmbeddingGenerator = generator });

var collection = vectorStore.GetCollection<string, FinanceInfo>("finances");
await collection.CreateCollectionAsync();


// Create some test data.
string[] FinanceInfo =
{
    "The budget for 2020 is EUR 100 000",
    "The budget for 2021 is EUR 120 000",
    "The budget for 2022 is EUR 150 000",
    "The budget for 2023 is EUR 200 000",
    "The budget for 2024 is EUR 364 000"
};

// Embeddings are generated automatically on upsert.
var records = FinanceInfo.Select((input, index) => new FinanceInfo { Key = index.ToString(), Text = input });
await collection.UpsertAsync(records);

// Embeddings for the search is automatically generated on search.
var searchResult = collection.SearchAsync(
    "What is my budget for 2024?",
    top: 1);

// Output the matching result.
await foreach (var result in searchResult)
{
    result.Display();
}

#r "nuget:Microsoft.Extensions.DependencyInjection"
#r "nuget:Microsoft.SemanticKernel.Connectors.Qdrant,*-*"

using Microsoft.Extensions.DependencyInjection;
using Microsoft.SemanticKernel;
using Microsoft.Extensions.AI;
using System.ClientModel;
using Microsoft.Extensions.AI;
using Azure.AI.OpenAI;
using Qdrant.Client;


var endpoint = "YOUR_AZURE_OPENAI_ENDPOINT";
var apikey = "YOUR_API_KEY";

var azureClinet = new AzureOpenAIClient(
    endpoint: new Uri(endpoint),
    credential: new ApiKeyCredential("apikey")
);

var services = new ServiceCollection();

var embeddingGenerator = azureClinet.GetEmbeddingClient("text-embedding-3-small")
                            .AsIEmbeddingGenerator();     


// Register embedding generator (依赖Microsoft.Extensions.AI NuGet 包)
services.AddEmbeddingGenerator(embeddingGenerator);

var qdrantEndpoint = "YOUR_QDRANT_ENDPOINT";
var qdrantApiKey= "YOUR_QDRANT_API_KEY";

var qdrantClient = new QdrantClient(
  address: new Uri("qdrantEndpoint"),
  apiKey: "qdrantApiKey"
);

services.AddSingleton<QdrantClient>(sp => qdrantClient); // 假设 qdrantClient 已经定义

var options = new QdrantVectorStoreOptions 
{
    EmbeddingGenerator = embeddingGenerator
};

// Register the Qdrant vector store
services.AddQdrantVectorStore(options); // 假设 AddQdrantVectorStore 已经定义

// Register the Qdrant vector store collection
services.AddQdrantVectorStoreRecordCollection<Guid, FinanceInfo>( 
    "finances",
    new QdrantVectorStoreRecordCollectionOptions<FinanceInfo>
    {
        EmbeddingGenerator = embeddingGenerator
    });


var serviceProvider = services.BuildServiceProvider();

var vectorStore = serviceProvider.GetRequiredService<IVectorStore>();

// List all collections
var collections = vectorStore.ListCollectionNamesAsync();
await foreach (var collectionName in collections)
{
    collectionName.Display();
}

// Get the collection 
// Note that the collection is created automatically if it does not exist
var collection = vectorStore.GetCollection<Guid, FinanceInfo>("finances"); 
collection.Display();

var embeddingGenerator = serviceProvider.GetRequiredService<IEmbeddingGenerator<string,Embedding<float>>>();
var embeddingResult = await embeddingGenerator.GenerateAsync("what's the weather like today?");
embeddingResult.Display();

var financeSearch = serviceProvider.GetRequiredService<IVectorSearch<FinanceInfo>>(); 
var searchResult = financeSearch.SearchAsync("What is my budget for 2024?", top: 1);
await foreach (var item in searchResult)
{
    item.Display();
}

// directly get the regirsted collection
var financeCollection = serviceProvider.GetRequiredService<IVectorStoreRecordCollection<Guid,FinanceInfo>>(); 

financeCollection.Display();

var result = financeCollection.SearchAsync("What is my budget for 2024?", top: 1);
await foreach (var item in result)
{
    item.Display();
}