PDF RAG

A Python package for building RAG (Retrieval-Augmented Generation) applications using PDFs, ChromaDB, and Ollama.

Project Structure

.
├── pdf_rag
│   ├── document_processor.py
│   ├── __init__.py
│   ├── llm_interface.py
│   ├── main.py
│   └── vector_store.py
├── README.md
├── requirements.txt
├── setup.py
├── test_package.py
└── test.py

2 directories, 10 files

Installation

1. Create and activate a virtual environment:

   # Create a virtual environment
   python -m venv venv
   
   # Activate the virtual environment
   # On Windows
   venv\Scripts\activate
   # On Unix or MacOS
   source venv/bin/activate

2. Install the package:

   pip install -e .

3. Install Ollama on Linux:

   Follow the steps below to install Ollama on a Linux system.

   # Download the Ollama installer
   curl -fsSL https://ollama.com/install.sh | sh
   
   # Verify the installation
   ollama --version

4. Download models in Ollama:

   To download specific models such as llama3 and deepseek-R1, use the following commands:

   # Download the llama3 model
   ollama pull llama3
   
   # Download the deepseek-R1 model
   ollama pull deepseek-R1

base Usage

from pdf_rag import PDFRAGApplication

# Initialize the application
rag = PDFRAGApplication(model_name= "deepseek-r1")

# Load a PDF
rag.load_pdf("your_document.pdf")

# Query the system
response = rag.query("What is this document about?")
print(response)

Testing

Run the test.py script to see how the module works with ChromaDB:

import chromadb
chroma_client = chromadb.Client()

# switch `create_collection` to `get_or_create_collection` to avoid creating a new collection every time
collection = chroma_client.get_or_create_collection(name="my_collection")

# switch `add` to `upsert` to avoid adding the same documents every time
collection.upsert(
    documents=[
        "This is a document about pineapple",
        "This is a document about oranges"
    ],
    ids=["id1", "id2"]
)

results = collection.query(
    query_texts=["This is a query document about hawaii"], # Chroma will embed this for you
    n_results=2 # how many results to return
)

print(results)