Installation

Requirements

• Python 3.13 or later

• Windows, Linux or macOS (Apple Silicon)

Install from PyPI

The recommended way to install pycanha is with pip:

pip install pycanha

This will automatically install pycanha-core (the compiled C++ backend) and matplotlib as dependencies.

Optional: Intel MKL acceleration

By default, pycanha uses MKL in Windows and Linux versions. If you don’t want to use MKL, the option is available, but you will need to build pycanha-core from source. MacOS version is shipped without MKL.

Development install

To install for development (editable mode with dev tools):

git clone https://github.com/pycanha-project/pycanha.git
cd pycanha
pip install -e .                 # Install pycanha in editable mode
pip install -e ".[doc]"          # For documentation dependencies
pip install -e ".[dev]"          # For development dependencies

Verifying the installation

import pycanha as pc
import pycanha.tmm as pm
import pycanha_core

pycanha_core.print_package_info()

tm = pc.ThermalModel("test")
print(f"Model name: {tm.name}")