Installation
Requirements
Specular Differentiation requires Python 3.14 or later and NumPy 2.4 or later. NumPy is installed automatically with the package and is the default backend.
Install from PyPI
Install Specular Differentiation using pip:
pip install specular-differentiation
The package is distributed as specular-differentiation and imported in
Python as specular.
The standard installation includes scaled_mean, derivative, gradient,
and jacobian, as well as the scalar ODE methods and the SPEG optimizer.
Verify the installation
Import specular and check the installed package version:
import specular
print("specular version:", specular.__version__)
Optional backends
Install the corresponding extra to use Numba or JAX:
pip install "specular-differentiation[numba]"
pip install "specular-differentiation[jax]"
The Numba extra requires numba >= 0.65. The JAX extra requires
jax >= 0.10 and jaxlib >= 0.10. pip installs these dependencies
automatically.
Installing an extra makes that backend available; NumPy remains the default. See Backends for backend selection and precision settings.
Development and documentation
For development, run the following command from the root of a local checkout of the repository:
pip install -e ".[dev]"
The editable installation uses the local source and includes both optional backends, the test dependencies, and the example tools.
Documentation tools can be installed separately from the same directory:
pip install -e ".[docs]"