Welcome to the Automotive Agent RAG repository. This project leverages Retrieval-Augmented Generation (RAG) to facilitate queries related to automotive manuals. By integrating intelligent agents, the system provides accurate and context-specific information about various automotive models.
- Retrieval-Augmented Generation (RAG): uses RAG techniques to enhance the accuracy of information retrieval.
- Automotive-Specific Queries: tailored for handling queries related to different car makes, models, and years.
- Intelligent Agents: employs advanced AI agents to understand and process user requests effectively.
- Milvus Integration: utilizes Milvus for efficient vector similarity search and retrieval.
Before you begin, ensure you have met the following requirements:
- Python 3.10 or later
- Jupyter Notebook
- Docker (if using Milvus with Docker)
- An OpenAI API key
- Required Python libraries (see
requirements.txt)
- Clone this repository:
git clone https://github.com/thaisaraujom/automotive-agent-rag.git
- Navigate to the project directory:
cd automotive-agent-rag - Install the required packages:
pip install -r requirements.txt
You have the option to set up Milvus either locally or using Docker. For detailed instructions on both methods, please refer to the Milvus Documentation.
If you do not have an OpenAI API key, you can sign up for an account on the OpenAI website and generate an API key.
To run the project, follow these steps:
- Open the
agent_rag_manual.ipynbfile in you Jupyter Notebook environment. - Follow the instructions within the notebook to execute the cells and interact with the system.
agent_rag_manual.ipynb: main Jupyter Notebook containing the implementation.requirements.txt: list of dependencies required to run the project.
This project is licensed under the MIT License - see the LICENSE file for details.