Usage

Quick Start

To begin using neuroptimiser, follow these steps:

  1. Clone the Repository:

    git clone https://github.com/neuroptimiser/neuroptimiser.git
cd neuroptimiser

  2. Create and Activate a Virtual Environment (Python 3.10 recommended):

    python3.10 -m venv .venv
source .venv/bin/activate  # On Unix/Linux/macOS
# .venv\Scripts\activate  # On Windows

  3. Install Dependencies:

    Using pip:

    pip install -e .
# or from PyPI
pip install neuroptimiser

    Using uv (recommended):

    uv pip install -e .
# or from PyPI
uv pip install neuroptimiser

    You can also use the provided Makefile for additional installation options and commands.

  4. Prevent “Too many open files” error (if needed):

    On Unix/Linux/macOS:

    ulimit -n 65536

    On Windows, this is typically not needed as Windows has higher default limits.

Example Usage

Here’s a simple example to get started:

from neuroptimiser import NeurOptimiser
import numpy as np

problem_function = lambda x: np.linalg.norm(x)
problem_bounds = np.array([[-5.0, 5.0], [-5.0, 5.0]])

optimiser = NeurOptimiser()

optimiser.solve(
    obj_func=problem_function,
    search_space=problem_bounds,
    debug_mode=True,
    num_iterations=1000,
)

Examples

Explore the following examples to understand how to utilise neuroptimiser:

Further Reading

For more detailed information, refer to the following sections: