graphfleet 0.2.3

Build upon GraphRAG, provide a quick solution setup and is part of the overall Qredence System like AgenticFleet and FleetUI

Overview

GraphFleet is an advanced implementation of GraphRAG from Microsoft, designed to enhance large language models' ability to reason about complex information and private datasets. It builds upon GraphRAG (Retrieval Augmented Generation using Graph structures) and will gradually adopt its own path to fulfill our roadmap at Qredence.

GraphFleet

GraphFleet uses knowledge graphs to provide substantial improvements in question-and-answer performance when reasoning about complex information. It addresses limitations of traditional RAG approaches by:

• Connecting disparate pieces of information through shared attributes.
• Structured, hierarchical approach to Retrieval Augmented Generation.
• Knowledge graph extraction from raw text.
• Community hierarchy building.
• Hierarchical summarization.
• Enhanced reasoning capabilities for LLMs on private datasets.
• Improve the structure of the overall repository.
• Add dedicated prompts for indexing and querying, and more vector databases.
• Add more notebooks.
• Provide a FleetUI Design Kit and a quicker way of starting GraphFleet locally.
• Add integrations (Langchain, Flowise, Langflow, Microsoft Fabric, Composio, Neo4j, etc.).
• Access GraphFleet through a secure and enterprise-ready Azure Cloud hosting version. Join the waitlist now.
• And way more... 👀

Key Features

• Structured, hierarchical approach to Retrieval Augmented Generation.
• Knowledge graph extraction from raw text.
• Community hierarchy building.
• Hierarchical summarization.
• Enhanced reasoning capabilities for LLMs on private datasets.

Getting Started

Prerequisites

• Python 3.10 (ideally 3.12)
• Poetry (for dependency management)

Installation

1. Clone the repository:

   </code></pre>
   </li>
   </ol>
   <p>git clone <a href="https://github.com/Qredence/GraphFleet.git">https://github.com/Qredence/GraphFleet.git</a>
   cd GraphFleet</p>
   <pre><code>
   
   2. Install dependencies using Poetry:
   

   poetry shell

   poetry install

   
   ### Usage
   
   1. Setting up your settings:
   
   

   cd graphfleet

   
   - Set up the necessary environment variables in the `.env.example` file and change the name to .env.
   The required ones being :
   
   

   Base LLM Settings

   GRAPHRAG_API_KEY="your_api_key" GRAPHRAG_API_BASE="http://.openai.azure.com" # For Azure OpenAI Users GRAPHRAG_API_VERSION="api_version" # For Azure OpenAI Users

   Text Generation Settings

   GRAPHRAG_LLM_TYPE="azure_openai_chat" # or openai_chat GRAPHRAG_LLM_DEPLOYMENT_NAME="gpt-4-turbo-preview" GRAPHRAG_LLM_MODEL_SUPPORTS_JSON=True

   Text Embedding Settings

   GRAPHRAG_EMBEDDING_TYPE="azure_openai_embedding" # or openai_embedding GRAPHRAG_LLM_DEPLOYMENT_NAME="text-embedding-3-small"

   Data Mapping Settings

   GRAPHRAG_INPUT_TYPE="text"

   
   
   - Open `settings.yaml` and fill the parameter you wish to fill according to your needs.
   
   
   2. Run the indexing process:
   
   [Get started quickly](notebook/get-started-graphfleet.ipynb)
   
   
   3. Perform queries in local mode or global mode depending on your usecase learn more in the [GraphRAG documentation](https://microsoft.github.io/graphrag/posts/query/overview/):
   For local query mode :
   
   

   python -m graphrag.query --root ./graphfleet --method local "What are the key features of GraphRAG ??"

   
   For Global query mode :
   
   

   python -m graphrag.query --root ./graphfleet --method global "What are the top main features of GraphRAG"

   ## Star History
   
   [![Star History Chart](https://api.star-history.com/svg?repos=Qredence/GraphFleet&type=Date)](https://star-history.com/#Qredence/GraphFleet&Date)