EVOSEAL User Manual¶

Welcome to the EVOSEAL User Manual. This document provides comprehensive information about using EVOSEAL effectively.

Table of Contents¶

Installation
Configuration
Basic Usage
Advanced Features
Troubleshooting
FAQs

Installation¶

Prerequisites¶

Python 3.10 or higher
pip (Python package manager)
Git

Installation Steps¶

Clone the Repository

git clone https://github.com/SHA888/EVOSEAL.git
cd EVOSEAL

Set Up Virtual Environment

python -m venv venv
source .venv/bin/activate  # On Windows: venv\Scripts\activate

Install Dependencies
```
pip install -r requirements.txt
```
Install in Development Mode
```
pip install -e .
```

Project Structure¶

EVOSEAL follows a modular structure:

evoseal/
├── core/               # Core framework components
├── integration/        # Integration modules (DGM, OpenEvolve, SEAL (Self-Adapting Language Models))
├── models/            # Data models
├── providers/         # AI/ML model providers
├── storage/           # Data persistence
├── utils/             # Utility functions
└── examples/          # Example scripts and templates
    ├── basic/        # Basic usage examples
    ├── workflows/    # Workflow examples
    └── templates/   # Project templates

Basic Usage¶

Running Examples¶

EVOSEAL provides several example scripts to help you get started:

Quickstart Example

python -m evoseal.examples.basic.quickstart

Logging Example

python -m evoseal.examples.basic.logging_example

Basic Usage Example

python -m evoseal.examples.basic.basic_usage

Using Project Templates¶

Start a new project using our template:

# Copy the template to a new directory
cp -r evoseal/examples/templates/basic my_project
cd my_project

# Install dependencies
pip install -r requirements.txt

Configuration¶

Create a .env file in your project root with the following variables:

# Required
OPENAI_API_KEY=your_openai_api_key
ANTHROPIC_API_KEY=your_anthropic_api_key

# Optional
LOG_LEVEL=INFO
CACHE_DIR=./.cache

Configuration Files¶

EVOSEAL supports multiple configuration files for different environments: - config/development.json - Development settings - config/testing.json - Testing settings - config/production.json - Production settings

Basic Usage¶

Initialization¶

from evoseal import EVOSEAL

# Initialize with default settings
evoseal = EVOSEAL()

Running Evolution¶

# Define your task
task = "Create a Python function that implements quicksort"

# Run evolution
result = evoseal.evolve(
    task=task,
    max_iterations=50,
    population_size=10
)

# Access results
print(f"Best solution: {result.best_solution}")
print(f"Fitness score: {result.fitness}")
print(f"Iterations completed: {result.iterations}")

Advanced Features¶

Custom Fitness Functions¶

def custom_fitness(solution):
    """Evaluate a solution based on specific criteria."""
    score = 0

    # Example: Reward shorter solutions
    score += max(0, 10 - len(solution) / 100)

    # Add your custom evaluation logic here

    return score

# Initialize with custom fitness
custom_evoseal = EVOSEAL(fitness_function=custom_fitness)

Model Selection¶

from evoseal.models import OpenAIModel, AnthropicModel

# Use specific models
gpt4 = OpenAIModel(model="gpt-4")
claude = AnthropicModel(model="claude-3-opus")

# Initialize with custom model
evoseal = EVOSEAL(model=gpt4)

Checkpointing¶

# Save checkpoint
evoseal.save_checkpoint("checkpoint.pkl")

# Load checkpoint
evoseal.load_checkpoint("checkpoint.pkl")

Troubleshooting¶

Common Issues¶

API Key Errors
Ensure your API keys are set in the .env file
Verify the keys have the correct permissions
Installation Issues
Make sure you're using Python 3.10 or higher
Try recreating your virtual environment
Performance Problems
Reduce population size or number of iterations
Use smaller models for faster iteration

Frequently Asked Questions¶

How do I improve evolution results?¶

Provide clear, specific tasks
Experiment with different population sizes
Adjust the number of iterations
Fine-tune the fitness function

Can I use my own models?¶

Yes! EVOSEAL supports custom model implementations. See the API reference for details.

How do I contribute?¶

Please see our Contribution Guidelines.

Support¶

For additional help, please open an issue on GitHub.

Last update: 2025-07-20
Created: 2025-06-17