Important Points:

  1. Deploying a Model in Azure AI Foundry:

    • Sign in to the Azure AI Foundry portal.
    • Search for and select the gpt-4o model.
    • Create a project with customized settings (resource name, subscription, resource group, region).
    • The project includes connections to Azure AI services and models.

  2. Creating a Client Application:

    • Use Azure AI Foundry and Azure AI Model Inference SDKs to develop an application.
    • Choose between Python or C# for development.

  3. Application Configuration:

    • Note the Azure AI Foundry project endpoint.
    • Clone the GitHub repo containing the code files.
    • Navigate to the folder containing the chat application code files.

  4. Writing Code to Connect and Chat:

    • Add references to necessary SDK namespaces.
    • Initialize the project client and chat client.
    • Create a loop to allow user input and retrieve completions from the model.

  5. Running the Application:

    • Sign into Azure and run the app.
    • Interact with the generative AI model by asking questions.

  6. Cleaning Up:

    • Delete the resource group to avoid unnecessary costs.

Code References:

  • Cloning GitHub Repo:

    rm -r mslearn-ai-foundry -f
git clone https://github.com/microsoftlearning/mslearn-ai-studio mslearn-ai-foundry

  • Navigating to Code Folder:

    cd mslearn-ai-foundry/labfiles/chat-app/python
cd mslearn-ai-foundry/labfiles/chat-app/c-sharp

  • Installing Libraries:

    python -m venv labenv
./labenv/bin/Activate.ps1
pip install python-dotenv azure-identity azure-ai-projects azure-ai-inference

dotnet add package Azure.Identity
dotnet add package Azure.AI.Projects --version 1.0.0-beta.9
dotnet add package Azure.AI.Inference --version 1.0.0-beta.5

  • Editing Configuration File:

    code .env
code appsettings.json

  • Adding References:

    from dotenv import load_dotenv
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient
from azure.ai.inference.models import SystemMessage, UserMessage, AssistantMessage

    using Azure.Identity;
using Azure.AI.Projects;
using Azure.AI.Inference;

  • Initializing Project Client:

    projectClient = AIProjectClient(
    credential=DefaultAzureCredential(exclude_environment_credential=True, exclude_managed_identity_credential=True),
    endpoint=project_connection,
)

    DefaultAzureCredentialOptions options = new() { ExcludeEnvironmentCredential = true, ExcludeManagedIdentityCredential = true };
var projectClient = new AIProjectClient(new Uri(project_connection), new DefaultAzureCredential(options));

  • Creating Chat Client:

    chat = projectClient.inference.get_chat_completions_client()

    ChatCompletionsClient chat = projectClient.GetChatCompletionsClient();

  • Initializing Prompt:

    prompt = [SystemMessage("You are a helpful AI assistant that answers questions.")]

    var prompt = new List<ChatRequestMessage>() { new ChatRequestSystemMessage("You are a helpful AI assistant that answers questions.") };

  • Getting Chat Completion:

    prompt.append(UserMessage(input_text))
response = chat.complete(model=model_deployment, messages=prompt)
completion = response.choices[0].message.content
print(completion)
prompt.append(AssistantMessage(completion))

    prompt.Add(new ChatRequestUserMessage(input_text));
var requestOptions = new ChatCompletionsOptions() { Model = model_deployment, Messages = prompt };
Response<ChatCompletions> response = chat.Complete(requestOptions);
var completion = response.Value.Content;
Console.WriteLine(completion);
prompt.Add(new ChatRequestAssistantMessage(completion));

  • Running the Application:

    az login
python chat-app.py
dotnet run

Related Posts