Skip to content

SETUP_GUIDE.md

Latest commit

 

History

History
214 lines (143 loc) · 3.69 KB

SETUP_GUIDE.md

File metadata and controls

214 lines (143 loc) · 3.69 KB
Raw

LangChain Tutorial - Setup Guide

Quick Setup (5 minutes)

1. Install Dependencies

# Clone the repository
git clone <your-repo-url>
cd langchain_tutorial

# Install packages
uv sync
# OR
pip install -e .

2. Configure Environment

Create .env file in project root:

# Required for OpenAI models (optional)
OPENAI_API_KEY=sk-...
OPENAI_LLM=gpt-4o-mini

# Required for Ollama (local, recommended)
OLLAMA_LLM=llama3.2:3b

# Optional: Other providers
GOOGLE_API_KEY=...
ANTHROPIC_API_KEY=...
TAVILY_API_KEY=...

3. Pre-download AI Models (Recommended)

For production or offline use:

python scripts/setup_models.py

This downloads:

  • HuggingFace Embeddings: BAAI/bge-large-en-v1.5 (~1.34 GB)

Benefits:

  • ✅ No 5-15 minute wait on first run
  • ✅ Works offline
  • ✅ Predictable deployment

Skip if:

  • Just experimenting (auto-downloads on first use)
  • Fast internet available

Advanced PDF Processing Setup

Required for rag_pdf_advanced.py

Install Tesseract OCR

Windows:

  1. Download: https://github.com/UB-Mannheim/tesseract/wiki
  2. Install to: C:\Program Files\Tesseract-OCR
  3. Add to PATH
  4. Restart terminal

macOS:

brew install tesseract

Linux:

sudo apt-get install tesseract-ocr

Verify Installation

tesseract --version

Testing Your Setup

Test Basic Agent

cd langchain-crash-course/5_agents_tools
python agent_tools_basic.py

Test PDF RAG (Basic)

cd langchain-crash-course/5_agents_tools
python rag_pdf.py
python agent_react_rag_pdf.py

Test Advanced PDF RAG

cd langchain-crash-course/5_agents_tools

# 1. Pre-download models (if not done)
python ../../scripts/setup_models.py

# 2. Add a PDF to test
cp /path/to/your.pdf pdfs/

# 3. Initialize vector store
python rag_pdf_advanced.py

# 4. Run agent
python agent_react_rag_pdf_advanced.py

Troubleshooting

"Model not found" Error

# Manually download the model
python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('BAAI/bge-large-en-v1.5')"

"Tesseract not found" Error

  • Ensure Tesseract is installed
  • Check it's in your PATH: tesseract --version
  • Restart terminal/IDE after installation

"UnstructuredPDFLoader failed" Warning

  • This is normal - system falls back to PyPDFLoader
  • Check logs for details
  • Ensure unstructured package is installed

Out of Memory

Use a smaller embedding model in rag_pdf_advanced.py:

# Replace in create_multimodal_embeddings():
model_name = "BAAI/bge-small-en-v1.5"  # 133 MB instead of 1.34 GB

Docker Setup (Optional)

Create Dockerfile:

FROM python:3.11-slim

# Install system dependencies
RUN apt-get update && apt-get install -y \
    tesseract-ocr \
    && rm -rf /var/lib/apt/lists/*

WORKDIR /app

# Copy and install Python dependencies
COPY pyproject.toml uv.lock ./
RUN pip install uv && uv sync

# Pre-download models
COPY scripts/setup_models.py scripts/
RUN python scripts/setup_models.py

# Copy application
COPY . .

CMD ["python", "langchain-crash-course/5_agents_tools/agent_react_rag_pdf_advanced.py"]

Build and run:

docker build -t langchain-tutorial .
docker run -it --env-file .env langchain-tutorial

Next Steps

  1. ✅ Complete setup steps above
  2. 📖 Read langchain-crash-course/5_agents_tools/README.md
  3. 🚀 Try the examples in order:
    • Basic agent → Basic RAG → Advanced RAG
  4. 🔧 Customize for your use case

Support

  • Documentation: See README files in each directory
  • Issues: Check troubleshooting sections
  • Examples: All scripts include inline documentation