// Usual setup: importing Semantic Kernel SDK and SkiaSharp, used to display images inline.

#r "nuget: Microsoft.SemanticKernel, 1.23.0"
#r "nuget: System.Numerics.Tensors, 8.0.0"
#r "nuget: SkiaSharp, 2.88.3"

#!import config/Settings.cs
#!import config/Utils.cs
#!import config/SkiaUtils.cs

using Microsoft.SemanticKernel;
using Microsoft.SemanticKernel.TextToImage;
using Microsoft.SemanticKernel.Embeddings;
using Microsoft.SemanticKernel.Connectors.OpenAI;
using System.Numerics.Tensors;

using Kernel = Microsoft.SemanticKernel.Kernel;

#pragma warning disable SKEXP0001, SKEXP0010

// Load OpenAI credentials from config/settings.json
var (useAzureOpenAI, model, azureEndpoint, apiKey, orgId) = Settings.LoadFromFile();

// Configure the three AI features: text embedding (using Ada), chat completion, image generation (DALL-E 3)
var builder = Kernel.CreateBuilder();

if(useAzureOpenAI)
{
    builder.AddAzureOpenAITextEmbeddingGeneration("text-embedding-ada-002", azureEndpoint, apiKey);
    builder.AddAzureOpenAIChatCompletion(model, azureEndpoint, apiKey);
    builder.AddAzureOpenAITextToImage("dall-e-3", azureEndpoint, apiKey);
}
else
{
    builder.AddOpenAITextEmbeddingGeneration("text-embedding-ada-002", apiKey, orgId);
    builder.AddOpenAIChatCompletion(model, apiKey, orgId);
    builder.AddOpenAITextToImage(apiKey, orgId);
}
   
var kernel = builder.Build();

// Get AI service instance used to generate images
var dallE = kernel.GetRequiredService<ITextToImageService>();

// Get AI service instance used to extract embedding from a text
var textEmbedding = kernel.GetRequiredService<ITextEmbeddingGenerationService>();

#pragma warning disable SKEXP0001

var prompt = @"
Think about an artificial object correlated to number {{$input}}.
Describe the image with one detailed sentence. The description cannot contain numbers.";

var executionSettings = new OpenAIPromptExecutionSettings 
{
    MaxTokens = 256,
    Temperature = 1
};

// Create a semantic function that generate a random image description.
var genImgDescription = kernel.CreateFunctionFromPrompt(prompt, executionSettings);

var random = new Random().Next(0, 200);
var imageDescriptionResult = await kernel.InvokeAsync(genImgDescription, new() { ["input"] = random });
var imageDescription = imageDescriptionResult.ToString();

// Use DALL-E 3 to generate an image. OpenAI in this case returns a URL (though you can ask to return a base64 image)
var imageUrl = await dallE.GenerateImageAsync(imageDescription.Trim(), 1024, 1024);

await SkiaUtils.ShowImage(imageUrl, 1024, 1024);

// Prompt the user to guess what the image is
var guess = await InteractiveKernel.GetInputAsync("Describe the image in your words");

// Compare user guess with real description and calculate score
var origEmbedding = await textEmbedding.GenerateEmbeddingsAsync(new List<string> { imageDescription } );
var guessEmbedding = await textEmbedding.GenerateEmbeddingsAsync(new List<string> { guess } );
var similarity = TensorPrimitives.CosineSimilarity(origEmbedding.First().Span, guessEmbedding.First().Span);

Console.WriteLine($"Your description:\n{Utils.WordWrap(guess, 90)}\n");
Console.WriteLine($"Real description:\n{Utils.WordWrap(imageDescription.Trim(), 90)}\n");
Console.WriteLine($"Score: {similarity:0.00}\n\n");

//Uncomment this line to see the URL provided by OpenAI
//Console.WriteLine(imageUrl);