Contributing to Agentic Retrieval Strategy

We're excited that you're interested in contributing to the Agentic Retrieval Strategy project! This document outlines how to contribute effectively.

🚀 Getting Started

Prerequisites

Python 3.8+ (recommended: 3.12+)
Ollama installed and running
Git for version control

Development Setup

# Fork and clone the repository
git clone https://github.com/yourusername/agentic-retrieval-strategy.git
cd agentic-retrieval-strategy

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # Linux/Mac
# .venv\Scripts\activate   # Windows

# Install development dependencies
pip install -r requirements.txt
pip install pytest black flake8  # Development tools

# Run tests to ensure setup works
python -m pytest tests/

🎯 Ways to Contribute

1. Bug Reports

Use GitHub Issues with the "bug" label
Include system information (Python version, OS, Ollama version)
Provide minimal reproduction steps
Include error messages and logs

2. Feature Requests

Use GitHub Issues with the "enhancement" label
Describe the use case and expected behavior
Consider implementation complexity and project scope

3. Code Contributions

New Search Strategies

# In MultiStrategyRetriever class
def your_custom_strategy(self, query, documents):
    """
    Implement your custom retrieval strategy
    
    Args:
        query (str): Search query
        documents (list): Document corpus
        
    Returns:
        list: Ranked document results
    """
    # Your implementation here
    return ranked_results

New Evaluation Metrics

def custom_evaluation_metric(response, reference, context):
    """
    Implement custom evaluation logic
    
    Returns:
        float: Score between 0.0 and 1.0
    """
    # Your evaluation logic
    return score

New Model Support

# Add to MODELS dictionary in configuration
'YOUR_MODEL': {
    'api_url': 'http://localhost:11434/api/generate',
    'model_name': 'your-model-name',
    'description': 'Model description'
}

4. Documentation

Improve README clarity
Add code examples
Translate to other languages
Create tutorial content

5. Testing

Add unit tests for new features
Improve test coverage
Add integration tests
Performance benchmarking

📋 Development Guidelines

Code Style

Follow PEP 8 Python style guide
Use Black for code formatting: black .
Use flake8 for linting: flake8 .
Add type hints where appropriate

Commit Messages

Use conventional commit format:

type(scope): description

Examples:
feat(retrieval): add neural search strategy
fix(evaluation): correct confidence calculation
docs(readme): update installation instructions
test(phase3): add multi-strategy unit tests

Branch Naming

feature/description for new features
fix/description for bug fixes
docs/description for documentation
test/description for testing improvements

Pull Request Process

Create Feature Branch

git checkout -b feature/your-feature-name

Make Changes
- Write code following project conventions
- Add/update tests as needed
- Update documentation

Test Your Changes

python -m pytest tests/
python phase3_agentic_rag.py  # Integration test

Submit Pull Request
- Use descriptive title and description
- Reference related issues
- Include screenshots/results if applicable
- Request review from maintainers

🧪 Testing Guidelines

Unit Tests

# tests/test_retrieval_strategies.py
def test_bm25_strategy():
    retriever = MultiStrategyRetriever()
    results = retriever.bm25_search("query", documents)
    assert len(results) > 0
    assert all(isinstance(r, dict) for r in results)

Integration Tests

# tests/test_end_to_end.py
def test_phase3_evaluation():
    # Test complete Phase 3 pipeline
    results = run_phase3_evaluation(test_queries, test_corpus)
    assert 'mistral' in results
    assert 'phi4' in results

Performance Tests

Benchmark new features against baseline
Document performance impact
Include memory and time complexity analysis

📊 Research Contributions

Academic Research

Novel retrieval strategies
Improved confidence metrics
Comparative studies with additional models
Domain-specific adaptations

Industry Applications

Enterprise deployment guides
Integration with existing systems
Scalability improvements
Production optimizations

Data Contributions

Domain-specific datasets
Multilingual corpora
Synthetic evaluation data
Ground truth annotations

🎯 Priority Areas

High Priority

Additional LLM model support (GPT-4, Claude, Gemini)
Real-time evaluation dashboard
Distributed evaluation support
Advanced confidence algorithms

Medium Priority

Multimodal retrieval strategies
Graph-based knowledge representation
Few-shot strategy learning
Human evaluation integration

Low Priority

GUI interface
Cloud deployment automation
Mobile app integration
Blockchain-based evaluation

🤝 Community

Communication Channels

GitHub Issues: Technical discussions and bug reports
GitHub Discussions: General questions and ideas
Discord: Real-time community chat (coming soon)

Code of Conduct

We follow the Contributor Covenant code of conduct. Please be respectful and inclusive in all interactions.

Recognition

Contributors will be acknowledged in:

README.md contributors section
Release notes for significant contributions
Academic publications using the framework

📚 Resources

Learning Materials

External Resources

🚀 Release Process

Version Numbering

We use semantic versioning (MAJOR.MINOR.PATCH):

MAJOR: Breaking changes
MINOR: New features (backward compatible)
PATCH: Bug fixes

Release Checklist

Thank you for contributing to Agentic Retrieval Strategy! Together, we're advancing the state of RAG evaluation and agentic AI systems.

For questions about contributing, please open an issue or contact the maintainers.

FilesExpand file tree

CONTRIBUTING.md

Latest commit

History