DigiSquares-Managed Vectors Quickstart

DigiSquares-Managed Vectors is a hosted vector database that enables you to store unstructured text from documents and web pages for use with AI models through DigiSquares. Digisquares abstracts away the complexities of preparing text and automatically generates the required data for AI models to make decisions using your data.

You can include related information from DigiSquares-Managed Vectors to provide more context. This allows you to instruct AI models to use your own data when making decisions, such as responding to a customer with context from support articles.

The following guide explains how to add text to DigiSquares-Managed Vectors and use it with the Generate Text AI Action.

Introduction

Welcome to the DigiSquares-Managed Vectors quickstart guide. This documentation will help you understand how to use vector databases in your Digisquares apps to create advanced search and retrieval systems based on vector similarity. By leveraging the power of DigiSquares-Managed Vectors, you can enhance AI model interactions with your own contextual data.

Getting Started

System Requirements

Before you begin, ensure your system meets the following requirements:

• Operating System: Windows 10 or later, macOS 10.15 or later, or a modern Linux distribution
• RAM: Minimum 4 GB
• Disk Space: Minimum 500 MB of free space
• Browser: Latest version of Chrome, Firefox, Safari, or Edge

Installation

Follow these steps to set up your Digisquares environment:

1. Sign in to your Digisquares organization.
2. Ensure you have the necessary permissions to create and manage vector indices.
3. Navigate to the Resources tab in your Digisquares dashboard and select Digisquares Vectors.

Creating Your First Vector Index

Index Setup

1. Click Create > Vector Index in the DigiSquares-Managed Vectors section.
2. Enter a name for your index and specify the dimensionality of the vectors (e.g., 512 for a typical BERT model).
3. Choose a vector storage option (e.g., internal storage or an integrated vector database service).
4. Click Create to initialize the vector index.

Adding Text to Vectors

1. Click Add Vectors to begin adding text data to your index.

2. Upload a CSV file or use an API to import text data. Each entry should include an identifier and the text content.

3. For example, a CSV file might look like this:

   id,text
   1,"This is a sample document about the features of the VelocityX Pro keyboard."
   2,"Support article describing common issues with the VelocityX Pro keyboard keys."
   ...

4. Map the columns in your CSV to the appropriate fields in the vector index.

5. Click Import to add the text data to your index.

Using Vectors with AI Actions

1. Create a new app or open an existing app in DigiSquares.

2. Add a query to perform a vector search. Select Vector Search as the query type.

3. Configure the query to search your vector index. Specify the text to search for and the number of results to return.

4. For example, to search for vectors similar to a given customer query:

   {
     "index": "support_articles",
     "text": "keys not working on VelocityX Pro keyboard",
     "top_k": 5
   }

5. Save the query and name it vectorSearch.

6. Add a component to display the search results, such as a Table or List component.

7. Bind the data from the vectorSearch query to the component to display the retrieved items.

Best Practices

• Consistent Naming: Use consistent and descriptive names for your vector indices and fields.
• Data Normalization: Ensure that your vectors are normalized (e.g., unit vectors) to improve search accuracy.
• Batch Processing: When adding large amounts of text data, use batch processing to optimize performance.
• Version Control: Regularly commit your project to a version control system like Git to track changes and collaborate with team members.

FAQ

Q: Can I update text entries after they have been added to the index?
A: Yes, you can update or delete text entries in the index through the Digisquares interface or API.

Q: Is there a limit to the amount of text data I can store?
A: The limit depends on your storage configuration and the performance considerations of your vector database.

Q: Does Digisquares support different vector similarity measures?
A: Yes, Digisquares supports various similarity measures such as cosine similarity, Euclidean distance, and more.

Conclusion

Thank you for using DigiSquares-Managed Vectors. We hope this quickstart guide helps you implement powerful vector search and retrieval features in your apps. For more detailed guides and tutorials, visit our official documentation.

---

This content should help you get started with DigiSquares-Managed Vectors efficiently. If you need additional sections or further details, please let me know!