Metadata-Version: 2.4
Name: sciba
Version: 2026.3.28.1
Summary: Exposing scipy.special to numba.
Author-email: Jan Provaznik <jan@provaznik.pro>
License-Expression: LGPL-3.0
Project-URL: Repository, https://github.com/jan-provaznik/sciba
Requires-Python: >=3.10
Description-Content-Type: text/plain
License-File: LICENSE
Requires-Dist: numba>=0.64.0
Requires-Dist: scipy
Dynamic: license-file

Using scipy.special from numba has never been easier. 
All you have to do is install the sciba module and let it perform its magic.

Installation

  pip install sciba

Example use

  import numba
  import numpy as np
  import scipy.special as ss

  @numba.njit
  def test_sciba (x):
      return ss.eval_hermite(3, x)

  print(np.isclose(test_sciba(5.0), 940))

Remarks

  - This project draws heavy inspiration from the numba-scipy library, 
    which has been officially put on hold.

  - Adds support for complex-valued function calls on platforms with
    ctypes.c_{float|double}_complex support. This affects, for instance, the
    eval_jacobi function which exposes a complex-valued interface.

  - Tested with scipy@1.17.1 and numba@0.64.0 with python@3.14.3 on linux.

