⚠️ This repository is archived.
All content and ongoing development have been moved to the main eShopLite repository: https://aka.ms/eshoplite/repo.
Please use the new repository for the latest updates, issues, and contributions.
eShopLite - Semantic Search is a reference .NET application implementing an eCommerce site with Search features using Keyword Search and Semantic Search. This sample specifically demonstrates the integration of the DeepSeek-R1 model for enhanced semantic understanding and search capabilities.
The reference application is part of the Generative AI for Beginners .NET series, which aims to provide practical examples and resources for developers interested in generative AI.
- Features
- Architecture diagram
- Getting started
- Deploying to Azure
- Run solution
- Resources
- Video Recordings
- Guidance
- Resources
GitHub CodeSpaces: This project is designed to be opened in GitHub Codespaces as an easy way for anyone to deploy the solution entirely in the browser.
This is the eShopLite Aplication running, performing a Semantic Search using deepseek-r1 model:
The Aspire Dashboard to check the running services:
The Azure Resource Group with all the deployed services:
WIP - Coming soon!
The solution is in the ./src folder, the main solution is eShopLite-Aspire-DeepSeekR1.sln.
Once you've opened the project in Codespaces, or locally, you can deploy it to Azure.
From a Terminal window, open the folder with the clone of this repo and run the following commands.
-
Login to Azure:
azd auth login
-
Provision and deploy all the resources:
azd up
It will prompt you to provide an
azdenvironment name (like "eShopLite"), select a subscription from your Azure account, and select a location where the necessary models are available (like "eastus2" or "swedencentral"). -
When
azdhas finished deploying, you'll see the list of resources created in Azure and a set of URIs in the command output. -
Visit the store URI, and you should see the eShop Lite app! 🎉
Note: The deploy files are located in the ./src/eShopAppHost/infra/ folder. They are generated by the Aspire AppHost project.
-
Create a new Codespace using the
Codebutton at the top of the repository. -
The Codespace creation process can take a couple of minutes.
-
Once the Codespace is loaded, it should have all the necessary requirements to deploy the solution.
To run the project locally, you'll need to make sure the following tools are installed:
- .NET 9
- Git
- Azure Developer CLI (azd)
- Visual Studio Code or Visual Studio
- If using Visual Studio Code, install the C# Dev Kit
- .NET Aspire workload: Installed with the Visual Studio installer or the .NET CLI workload.
- An OCI compliant container runtime, such as:
- Docker Desktop or Podman.
Follow these steps to run the project, locally or in CodeSpaces:
-
Navigate to the Aspire Host folder project using the command:
cd ./src/eShopAppHost/ -
Set the user secrets for the Azure OpenAI connection string as explained in the Local development using existing models section.
-
Run the project:
dotnet run
In order to use existing models: deepseek-r1, gpt-4o-mini and text-embedding-ada-002, you need to define the specific connection string in the Products project.
Add a user secret named ConnectionStrings:openai with the configuration for the Azure OpenAI service that contains the gpt-4o-mini and text-embedding-ada-002 models, and a user secret named ConnectionStrings:deepseek-r1 with the configuration for the Azure OpenAI service that contains the DeepSeek-R1 model.
cd src/Products
dotnet user-secrets set "ConnectionStrings:openai" "Endpoint=https://<endpoint>.openai.azure.com/;Key=<key>;"
dotnet user-secrets set "ConnectionStrings:deepseek-r1" "Endpoint=https://<endpoint>.openai.azure.com/;Key=<key>;"This Azure OpenAI services must contain:
- a
deepseek-r1model named DeepSeek-R1 - a
gpt-4o-minimodel named gpt-4o-mini - a
text-embedding-ada-002model named text-embedding-ada-002
To use these services, the program.cs will create a client in this way:
var azureOpenAiClientName = "openai";
builder.AddAzureOpenAIClient(azureOpenAiClientName);This sample demonstrates the use of the DeepSeek-R1 model for semantic search capabilities. DeepSeek-R1 is a powerful large language model that enhances the search experience by:
- Improving natural language understanding for more accurate search results
- Delivering enhanced context awareness for better product matching
- Processing complex search queries with greater precision
- Providing high-quality semantic embeddings for product data
The implementation showcases how modern AI models like DeepSeek-R1 can significantly improve e-commerce search experiences by understanding user intent beyond simple keyword matching.
The eShopLite solution leverages the Aspire Dashboard and Azure Application Insights to provide comprehensive telemetry and monitoring capabilities
The .NET Aspire Dashboard offers a centralized view of the application's performance, health, and usage metrics. It integrates seamlessly with the Azure OpenAI services, allowing developers to monitor the performance of the gpt-4o-mini and text-embedding-ada-002 models. The dashboard provides real-time insights into the application's behavior, helping to identify and resolve issues quickly.
Azure Application Insights complements the Aspire Dashboard by offering deep diagnostic capabilities and advanced analytics. It collects detailed telemetry data, including request rates, response times, and failure rates, enabling developers to understand how the application is performing under different conditions. Application Insights also provides powerful querying and visualization tools, making it easier to analyze trends and detect anomalies.
By combining the Aspire Dashboard with Azure Application Insights, the eShopLite solution ensures robust monitoring and diagnostics, enhancing the overall reliability and performance of the application.
For Azure OpenAI Services, pricing varies per region and usage, so it isn't possible to predict exact costs for your usage. The majority of the Azure resources used in this infrastructure are on usage-based pricing tiers. However, Azure Container Registry has a fixed cost per registry per day.
You can try the Azure pricing calculator for the resources:
- Azure OpenAI Service: S0 tier, gpt-4o-mini and text-embedding-ada-002 models. Pricing is based on token count. Pricing
- Azure Container App: Consumption tier with 0.5 CPU, 1GiB memory/storage. Pricing is based on resource allocation, and each month allows for a certain amount of free usage. Pricing
- Azure Container Registry: Basic tier. Pricing
- Log analytics: Pay-as-you-go tier. Costs based on data ingested. Pricing
- Azure Application Insights pricing is based on a Pay-As-You-Go model. Pricing.
azd down.
Samples in this templates uses Azure OpenAI Services with ApiKey and Managed Identity for authenticating to the Azure OpenAI service.
The Main Sample uses Managed Identity](https://learn.microsoft.com/entra/identity/managed-identities-azure-resources/overview) for authenticating to the Azure OpenAI service.
Additionally, we have added a GitHub Action that scans the infrastructure-as-code files and generates a report containing any detected issues. To ensure continued best practices in your own repository, we recommend that anyone creating solutions based on our templates ensure that the Github secret scanning setting is enabled.
You may want to consider additional security measures, such as:
- Protecting the Azure Container Apps instance with a firewall and/or Virtual Network.
