graphevol 0.0.1

Evolutionary graph-based optimization library with Python bindings

graphevol

A scalable topology-evolution engine powered by a high-performance Rust backend.

graphevol provides fast graph-based neuroevolution in Python, with structural mutations, pruning, configurable operations, and a JAX/PyTorch/NumPy-friendly workflow.

---

Features

• Fast Rust backend for graph topology evolution
• Six structural mutation operators
• Configurable GA via YAML (population, mutation rates, pruning, crossover, etc.)
• Customizable unary & binary operations (ops.yaml)
• Clean Python API for research workflows
• Good performance with JAX, NumPy, PyTorch

---

Installation

pip install graphevol

(Optional) Install Graphviz for visualization:

Linux

sudo apt install graphviz

macOS

brew install graphviz

---

Quick Start

import graphevol as ge

# Generate default YAML configs
ge.generate_default_settings("ga.yaml", "ops.yaml")

genus = ge.PyGenus(
    seed=0,
    input_dim=4,
    output_dim=2,
    ga_config_path="ga.yaml",
    ops_path="ops.yaml",
)

genomes = genus.next_generation() # weights and function's expression (str)
scores = rollout_fn(genomes)

genus.selection(scores)

---

⚙ Configuration

graphevol uses two YAML files:

ga_config.yaml

ops.yaml

---

Documentation

Full documentation & examples:
https://github.com/otyanokosaisai/graphevol

---

License

MIT License.

---

Contributions

Issues & PRs are welcome.