Using the MongoDB Vector Store connector (Preview)

More info coming soon.

Getting started

Add the MongoDB Vector Store connector NuGet package to your project.

dotnet add package Microsoft.SemanticKernel.Connectors.MongoDB --prerelease

You can add the vector store to the IServiceCollection dependency injection container using extension methods provided by Semantic Kernel.

using Microsoft.Extensions.DependencyInjection;

// Using IServiceCollection with ASP.NET Core.
var builder = WebApplication.CreateBuilder(args);
builder.Services.AddMongoVectorStore(connectionString, databaseName);

Extension methods that take no parameters are also provided. These require an instance of MongoDB.Driver.IMongoDatabase to be separately registered with the dependency injection container.

using Microsoft.Extensions.DependencyInjection;
using MongoDB.Driver;

// Using IServiceCollection with ASP.NET Core.
var builder = WebApplication.CreateBuilder(args);
builder.Services.AddSingleton<IMongoDatabase>(
    sp =>
    {
        var mongoClient = new MongoClient(connectionString);
        return mongoClient.GetDatabase(databaseName);
    });
builder.Services.AddMongoVectorStore();

You can construct a MongoDB Vector Store instance directly.

using Microsoft.SemanticKernel.Connectors.MongoDB;
using MongoDB.Driver;

var mongoClient = new MongoClient(connectionString);
var database = mongoClient.GetDatabase(databaseName);
var vectorStore = new MongoVectorStore(database);

It is possible to construct a direct reference to a named collection.

using Microsoft.SemanticKernel.Connectors.MongoDB;
using MongoDB.Driver;

var mongoClient = new MongoClient(connectionString);
var database = mongoClient.GetDatabase(databaseName);
var collection = new MongoCollection<string, Hotel>(
    database,
    "skhotels");

Data mapping

The MongoDB Vector Store connector provides a default mapper when mapping data from the data model to storage.

This mapper does a direct conversion of the list of properties on the data model to the fields in MongoDB and uses MongoDB.Bson.Serialization to convert to the storage schema. This means that usage of the MongoDB.Bson.Serialization.Attributes.BsonElement is supported if a different storage name to the data model property name is required. The only exception is the key of the record which is mapped to a database field named _id, since all MongoDB records must use this name for ids.

Property name override

For data properties and vector properties, you can provide override field names to use in storage that is different to the property names on the data model. This is not supported for keys, since a key has a fixed name in MongoDB.

The property name override is done by setting the BsonElement attribute on the data model properties.

Here is an example of a data model with BsonElement set.

using Microsoft.Extensions.VectorData;
using MongoDB.Bson.Serialization.Attributes;

public class Hotel
{
    [VectorStoreKey]
    public ulong HotelId { get; set; }

    [BsonElement("hotel_name")]
    [VectorStoreData(IsIndexed = true)]
    public string HotelName { get; set; }

    [BsonElement("hotel_description")]
    [VectorStoreData(IsFullTextIndexed = true)]
    public string Description { get; set; }

    [BsonElement("hotel_description_embedding")]
    [VectorStoreVector(4, DistanceFunction = DistanceFunction.CosineSimilarity)]
    public ReadOnlyMemory<float>? DescriptionEmbedding { get; set; }
}

Getting started

Add the MongoDB Atlas Vector Store dependencies to your environment. It needs the pymongo package which is included in the mongo extra: , you need to install with these extras:

pip install semantic-kernel[mongo]

You can then create the vector store.

from semantic_kernel.connectors.mongodb import MongoDBAtlasStore

# If the right environment settings are set, namely MONGODB_ATLAS_CONNECTION_STRING and optionally MONGODB_ATLAS_DATABASE_NAME and MONGODB_ATLAS_INDEX_NAME, this is enough to create the Store:
store = MongoDBAtlasStore()

Alternatively, you can also pass in your own mongodb client if you want to have more control over the client construction:

from pymongo import AsyncMongoClient
from semantic_kernel.connectors.mongodb import MongoDBAtlasStore

client = AsyncMongoClient(...)
store = MongoDBAtlasStore(mongo_client=client)

When a client is passed in, Semantic Kernel will not close the connection for you, so you need to ensure to close it, for instance with a async with statement.

You can also create a collection directly, without the store.

from semantic_kernel.connectors.mongodb import MongoDBAtlasCollection

# `hotel` is a class created with the @vectorstoremodel decorator
collection = MongoDBAtlasCollection(
    record_type=hotel,
    collection_name="my_collection",
)

Serialization

Since the MongoDB Atlas connector needs a simple dict with the fields corresponding to the index as the input, the serialization is quite easy, it only uses a predetermined key _id, so we replace the key of the data model with that if it is not already _id.

For more details on this concept see the serialization documentation.

Coming soon

More info coming soon.