ResearchAI 1.2.3

ResearchAI (.NET Class Library)

ResearchAI is a modular .NET class library for building AI-powered applications with Azure OpenAI, chat history management, and computer vision/image analytics. It is designed for easy integration into your .NET projects, supporting both in-memory and MongoDB storage, and providing extensible service interfaces.

Features

• Azure OpenAI Integration: Seamlessly connect to Azure OpenAI for chat completions and generative AI tasks.
• Chat History Management: Store, retrieve, and manage chat conversations using in-memory or MongoDB backends.
• Vision/Image Analytics: Analyze images and extract insights using Azure Computer Vision APIs.
• Extensible Architecture: Swap or extend storage and service implementations as needed.

Installation & Configuration

1. Add Reference

   • Add the ResearchAI library to your .NET project (via NuGet).

2. Configure Services

   • Register services in your DI container:

     string systemContextPath = Path.Combine(Directory.GetCurrentDirectory(), "system_context");
      string systemContext = File.ReadAllText(systemContextPath);

      builder.Services.AddAzureOpenAI(builder.Configuration, systemContext);

Usage Example: Web API Controller

Azure OpenAI Chat Completion

using Microsoft.AspNetCore.Mvc;
using ResearchAI.Services;

[ApiController]
[Route("api/[controller]")]
public class ChatController : ControllerBase
{
    private readonly AzureOpenAI _azureOpenAI;

    public ChatController(AzureOpenAI azureOpenAI)
    {
        _azureOpenAI = azureOpenAI;
    }

    // POST:/stream
    [HttpPost("stream")]
    public async Task StreamMessage([FromBody] CoversationRequest request)
    {

        Response.Headers.Append("Cache-Control", "no-cache");
        Response.Headers.Append("X-Accel-Buffering", "no");
        Response.ContentType = "text/event-stream";

        await foreach (var chunk in StreamMessagesAsync(SessionId, request.ConversationId, request.Message))
        {
            await Response.WriteAsync(chunk);
            await Response.Body.FlushAsync();
        }
    }

    private async IAsyncEnumerable<string> StreamMessagesAsync(string sessionId, string conversationId, string userInput)
    {
        await foreach (var response in _azureOpenAI.StreamMessageAsync(sessionId, conversationId, userInput))
        {
            yield return response;
        }
    }
}

ImageClient Example

using ResearchAI.Vision;

// Create an ImageClient instance
var imageClient = new ImageClient(new ImageClientModel
{
    Endpoint = "<your-vision-endpoint>",
    Key = "<your-vision-key>"
});

// Analyze an image (basic)
var analysisResult = await imageClient.AnalyzeImageAsync("https://example.com/image.jpg");

// If you want to use AzureOpenAI for advanced vision+chat scenarios:
var openAI = new AzureOpenAI(openAIOptions, null);
var imageClientWithAI = new ImageClient(new ImageClientModel
{
    Endpoint = "<your-vision-endpoint>",
    Key = "<your-vision-key>"
}, openAI);

var resultWithChat = await imageClientWithAI.AnalyzeImageWithChatAsync("https://example.com/image.jpg", "session-id");

License

This project is licensed under the terms of the Apache License Version 2.0.