GenAIStarter

GenAIStarter is a Visual Studio Code extension that helps you quickly generate Generative AI projects using Amazon Bedrock templates or LangChain.

Changelog

Version 1.5.1 (Latest)

- Added support for Hugging Face projects - Introduced new project types: Text Generation, Text Classification, Question Answering, Document Question Answering, Translation, Summarization, and Text-to-Image - Expanded project base selection to include Hugging Face

With the latest update, GenAIStarter now supports Hugging Face projects, allowing you to quickly set up and use a wide range of pre-trained models for various AI tasks.

Hugging Face Project Types

When creating a Hugging Face project, you can choose from the following project types:

• Text Generation
• Text Classification
• Question Answering
• Document Question Answering
• Translation
• Summarization
• Text-to-Image

Using Hugging Face Projects

1. In the project base selection, choose "Hugging Face".
2. Select the desired project type from the dropdown menu.
3. Once you select a project type, a list of relevant pre-trained models will appear.
4. Choose one or more models from the list by checking the corresponding boxes.
5. Click "Create" to generate your Hugging Face project.

The generated project will include sample code for both pipeline usage and direct model loading, making it easy to get started with your chosen models.

Notes for Hugging Face Projects

• Remember to handle API keys or model authentication as required by the specific models you choose.

1.5.0v Features

• Provides a command to create new Generative AI projects
• Supports creation of different project types:
  • AWS Bedrock: Lambda-based and Non-Lambda-based projects
  • LangChain projects with customizable dependencies
  • Hugging Face projects with various model types and pre-trained models (v2.0.0+)
• Offers an interactive webview interface for project configuration
• Integrates seamlessly with VS Code's command palette and workspace
• Supports dark/light mode toggle in the webview interface

Version 1.4.2

Features

• Provides a command to create new Generative AI projects
• Supports creation of different project types:
  • AWS Bedrock: Lambda-based and Non-Lambda-based projects
  • LangChain projects with customizable dependencies
• Offers an interactive webview interface for project configuration
• Integrates seamlessly with VS Code's command palette and workspace
• Supports dark/light mode toggle in the webview interface

Installation

1. Ensure you have Visual Studio Code 1.94.0 or higher installed
2. Open VS Code
3. Go to the Extensions view by clicking on the square icon in the left sidebar or pressing Ctrl+Shift+X (Windows/Linux) or Cmd+Shift+X (macOS)
4. Search for "GenAIStarter" in the Extensions view search bar
5. Click on the "Install" button next to the GenAIStarter extension

Usage

1. Open the Command Palette in VS Code:

   • On Windows/Linux: Press Ctrl+Shift+P
   • On macOS: Press Cmd+Shift+P

2. Type "GenAI Starter" in the Command Palette and select the command when it appears.
   

   
   
   

3. A new webview will open within VS Code, providing an interactive interface for configuring your project. Follow these steps in the webview:
   

   
   
   

   We can select dark mode as well:
   

   
   
   

   a. Select the project folder:

   • Click the "Select Folder" button
   • Choose a directory where you want to create your project
   • The selected folder path will appear in the input field
     

   
   
   
   

   b. Choose the project base:

   • Use the dropdown menu to select either "AWS Bedrock" or "LangChain"

c. For AWS Bedrock projects: - Select the project type: "Lambda Based" or "Non-Lambda Based"

For AWS Bedrock prject base , RAG utility creation available.

d. For LangChain projects: - Select the desired dependencies from the provided checkboxes

e. Click the "Create" button to generate your project

4. Wait for the project creation process to complete.

5. Once the project is created, it will automatically open in a new VS Code window.
   

   

   
   

   With RAG utility, it will create rag_util package with required files under it.

   

Additional Features

• Theme Toggle: Switch between light and dark modes for better visibility.
• Project Base Selection: Choose between AWS Bedrock and LangChain as the base for your project.
• AWS Bedrock Options: Select between Lambda-based and Non-Lambda-based projects for AWS Bedrock.
• LangChain Dependencies: Customize your LangChain project by selecting from various dependencies.
• Status Updates: Receive real-time status updates during the project creation process.

Troubleshooting

If you encounter any issues while using the extension:

1. Ensure you have the latest version of VS Code and the GenAIStarter extension installed.
2. If the problem persists, try reloading the VS Code window (Developer: Reload Window in the Command Palette) or restarting VS Code.

More Information

For more details on how to use the underlying bedrock-genai-builder tool, which powers the AWS Bedrock part of this extension, please refer to the following resource:

• How to use bedrock-genai-builder

This guide provides in-depth information about the capabilities and usage of the bedrock-genai-builder, which can help you better understand the AWS Bedrock projects generated by this extension.

For information about LangChain and its usage, please refer to the official LangChain documentation:

• LangChain Documentation

Enjoy building your Generative AI projects with GenAIStarter!