Installation¶

System Requirements¶

Python: 3.10, 3.11, 3.12, or 3.13
Operating System: Linux, macOS, or Windows
Hardware: CPU or NVIDIA GPU

Installation Options¶

The easiest way to get started with ORC is to install from PyPI:

pip install OpenReservoirComputing

If you're interested in the latest, unreleased version or in contributing, you can install from source. Please see the Contribution guidelines below for more details.

For CPU-only usage, clone the repository and install:

git clone https://github.com/Jan-Williams/OpenReservoirComputing.git
cd OpenReservoirComputing
pip install .

For GPU acceleration with CUDA support:

git clone https://github.com/Jan-Williams/OpenReservoirComputing.git
cd OpenReservoirComputing
pip install ".[gpu]"

Development Installation¶

For contributors or advanced users who want to modify the code:

git clone https://github.com/Jan-Williams/OpenReservoirComputing.git
cd OpenReservoirComputing
pip install -e ".[dev]"

This includes additional tools for testing, formatting, and documentation.

Complete Installation¶

To install all optional dependencies (GPU, development, notebooks, documentation):

pip install -e ".[all]"

Verification¶

Test your installation with this simple script:

import orc
import jax

# Check JAX device (CPU or GPU)
print(f"JAX is using: {jax.devices()}")

# Test basic ORC functionality
U, t = orc.data.lorenz63(tN=10, dt=0.01)
print(f"Generated Lorenz data shape: {U.shape}")

# Test model creation
esn = orc.forecaster.ESNForecaster(data_dim=3, res_dim=100)
print("ESN created successfully!")
print("Installation verified ✓")

Getting Help¶

If you encounter issues:

Check the GitHub Issues
Ensure your Python version is supported (3.10-3.13)
Try a clean virtual environment installation
For GPU issues, verify CUDA installation with nvidia-smi