Mirrors/SqlDatabaseVectorSearch
1
0
Fork 0
mirror of https://github.com/marcominerva/SqlDatabaseVectorSearch.git synced 2026-06-20 12:23:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

Improve README.md for clarity and consistency

Browse Source 
Updated the README.md file to enhance readability and grammatical structure. Key changes include improved phrasing, added missing commas, consistent formatting of section headers and bullet points, and correction of typographical errors. The description of the response streaming feature was also clarified to provide a better understanding of the application's functionality.
This commit is contained in:
Marco Minerva
2025-06-24 12:25:42 +02:00
parent 30fba5cfe0
commit 4e01ec81be
1 changed files with 17 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+17 -18
View File
@@ -1,9 +1,9 @@
# SQL Database Vector Search Sample # SQL Database Vector Search Sample
A repository that showcases the native VECTOR type in Azure SQL Database to perform embeddings and RAG with Azure OpenAI. A repository that showcases the native VECTOR type in Azure SQL Database to perform embeddings and RAG with Azure OpenAI.
The application allows to load documents, generate embeddings and save them into the database as Vectors, and perform searches using Vector Search and RAG. Currently, PDF, DOCX, TXT and MD files are supported. Vectors are saved and retrieved with Entity Framework Core using the [EFCore.SqlServer.VectorSearch](https://github.com/efcore/EfCore.SqlServer.VectorSearch) library. Embedding and Chat Completion are integrated with [Semantic Kernel](https://github.com/microsoft/semantic-kernel). The application allows you to load documents, generate embeddings, save them into the database as vectors, and perform searches using Vector Search and RAG. Currently, PDF, DOCX, TXT, and MD files are supported. Vectors are saved and retrieved with Entity Framework Core using the [EFCore.SqlServer.VectorSearch](https://github.com/efcore/EfCore.SqlServer.VectorSearch) library. Embedding and Chat Completion are integrated with [Semantic Kernel](https://github.com/microsoft/semantic-kernel).
This repository contains a Blazor Web App as well as a Minimal API that allows to programmatically interact with embeddings and RAG. This repository contains a Blazor Web App as well as a Minimal API that allows you to programmatically interact with embeddings and RAG.
### Web App ### Web App
![SQL Database Vector Search Web App](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/assets/SqlDatabaseVectorSearch_WebApp.png) ![SQL Database Vector Search Web App](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/assets/SqlDatabaseVectorSearch_WebApp.png)
@@ -14,19 +14,19 @@ This repository contains a Blazor Web App as well as a Minimal API that allows t
## Setup ## Setup
- [Create an Azure SQL Database](https://learn.microsoft.com/en-us/azure/azure-sql/database/single-database-create-quickstart) - [Create an Azure SQL Database](https://learn.microsoft.com/en-us/azure/azure-sql/database/single-database-create-quickstart)
- Open the [appsettings.json](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/appsettings.json) file and set the connection string to the database and the other settings required by Azure OpenAI - Open the [appsettings.json](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/appsettings.json) file and set the connection string to the database and the other settings required by Azure OpenAI.
- If your embedding model supports shortening, like **text-embedding-3-small** and **text-embedding-3-large**, and you want to use this feature, you need to set the [`Dimensions`](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/appsettings.json#L17) property to the corresponding value. If your model doesn't provide this feature, or do you want to use the default size, just leave the [`Dimensions`](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/appsettings.json#L17) property to NULL. Keep in mind that **text-embedding-3-small** has a dimension of 1536, while **text-embedding-3-large** uses vectors with 3072 elements, so with this latter model it is mandatory to specify a value (that must be less or equal to 1998, the maximum currently supported by the VECTOR type). - If your embedding model supports shortening, like **text-embedding-3-small** and **text-embedding-3-large**, and you want to use this feature, you need to set the [`Dimensions`](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/appsettings.json#L17) property to the corresponding value. If your model doesn't provide this feature, or if you want to use the default size, just leave the [`Dimensions`](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/appsettings.json#L17) property as NULL. Keep in mind that **text-embedding-3-small** has a dimension of 1536, while **text-embedding-3-large** uses vectors with 3072 elements, so with this latter model it is mandatory to specify a value (that must be less than or equal to 1998, the maximum currently supported by the VECTOR type).
- You may need to update the size of the [`VECTOR`](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/DataAccessLayer/ApplicationDbContext.cs?plain=1#L42C1-L42C47) column to match the size of the embedding model. The default value is 1536. Currently, the maximum allowed value is 1998. If you change it, remember to update also the [Database Migration](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/DataAccessLayer/Migrations/00000000000000_Initial.cs?plain=1#L35C1-L35C92). - You may need to update the size of the [`VECTOR`](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/DataAccessLayer/ApplicationDbContext.cs?plain=1#L42C1-L42C47) column to match the size of the embedding model. The default value is 1536. Currently, the maximum allowed value is 1998. If you change it, remember to also update the [Database Migration](https://github.com/marcominerva/SqlDatabaseVectorSearch/blob/master/SqlDatabaseVectorSearch/DataAccessLayer/Migrations/00000000000000_Initial.cs?plain=1#L35C1-L35C92).
- Run the application and start importing your documents - Run the application and start importing your documents.
- If you want to directly use the APIs: - If you want to directly use the APIs:
- import your documents with the `/api/documents` endpoint. - Import your documents with the `/api/documents` endpoint.
- Ask questions using `/api/ask` or `/api/ask-streaming` endpoints. - Ask questions using the `/api/ask` or `/api/ask-streaming` endpoints.
## Supported features ## Supported features
- **Conversation History with Question Reformulation**: This feature allows users to view the history of their conversations, including the ability to reformulate questions for better clarity and understanding. This ensures that users can track their interactions and refine their queries as needed. - **Conversation History with Question Reformulation**: This feature allows users to view the history of their conversations, including the ability to reformulate questions for better clarity and understanding. This ensures that users can track their interactions and refine their queries as needed.
- **Information about Token Usage**: Users can access detailed information about token usage, which helps in understanding the consumption of tokens during interactions. This feature provides transparency and helps users manage their token usage effectively. - **Information about Token Usage**: Users can access detailed information about token usage, which helps in understanding the consumption of tokens during interactions. This feature provides transparency and helps users manage their token usage effectively.
- **Response streaming**: This feature enables real-time streaming of responses, allowing users to receive information as it is being processed. This ensures a seamless and efficient flow of information, enhancing the overall user experience. - **Response Streaming**: This feature enables real-time streaming of responses, allowing users to receive information as it is being processed. This ensures a seamless and efficient flow of information, enhancing the overall user experience.
- **Citations**: The application provides citations for the sources used to justify each answer. This allows users to verify the information and understand the origin of the content provided by the system. - **Citations**: The application provides citations for the sources used to justify each answer. This allows users to verify the information and understand the origin of the content provided by the system.
### Example of JSON response ### Example of JSON response
@@ -73,7 +73,7 @@ This repository contains a Blazor Web App as well as a Minimal API that allows t
### How response streaming works ### How response streaming works
When using the `/api/ask-streaming` endpoint, answers will be streamed as happens with the typical response from OpenAI. The format of the response is the following: When using the `/api/ask-streaming` endpoint, answers will be streamed as with the typical response from OpenAI. The format of the response is as follows:
```json ```json
[ [
@@ -241,16 +241,15 @@ When using the `/api/ask-streaming` endpoint, answers will be streamed as happen
} }
] ]
``` ```
- The first piece of the response has the following characteristics: - The first piece of the response has the following characteristics:
- the *streamState* property is set to `Start` - The *streamState* property is set to `Start`.
- it contains the question and its reformulation (if not requested, *reformulatedQuestion* will be equals to *originalQuestion*) - It contains the question and its reformulation (if not requested, *reformulatedQuestion* will be equal to *originalQuestion*).
- the *tokenUsage* section holds information about token used for reformulation (if done) and for the embedding of the question - The *tokenUsage* section holds information about tokens used for reformulation (if done) and for the embedding of the question.
- Then, there are as many elements for the actual answer as necessary: - Then, there are as many elements for the actual answer as necessary:
- each one contains a token - Each one contains a token.
- The *streamState* property is set to `Append` - The *streamState* property is set to `Append`.
- *origianlQuestion*, *reformulatedQuestion* and *tokenUsage* are always `null` - *originalQuestion*, *reformulatedQuestion*, and *tokenUsage* are always `null`.
- The stream ends when an element with *streamState* equals to `End` is received. This element contains token usage information for the question and the whole answer and the list of citations. - The stream ends when an element with *streamState* equals `End` is received. This element contains token usage information for the question and the whole answer, and the list of citations.
> [!NOTE] > [!NOTE]
> If you prefer to use straight SQL, check out the [sql branch](https://github.com/marcominerva/SqlDatabaseVectorSearch/tree/sql). > If you prefer to use straight SQL, check out the [sql branch](https://github.com/marcominerva/SqlDatabaseVectorSearch/tree/sql).