Changelog
Stockwell
Python package for time-frequency analysis through Stockwell transform.
Based on original code from NIMH MEG Core Facility.
2021-2026 Claudio Satriano satriano@ipgp.fr
Unreleased
Add support for Python 3.14
Minimum required Python version: 3.9
Move tests outside the package directory
Validate inputs: reject empty arrays and non-integer parameters
Fix FFTW static plan caching to prevent segfault at exit
New tests: input validation, kazemi window, roundtrip, chirp example
Fix Hilbert phase: avoid shifting the Hilbert transform in frequency when multiplying by the Gaussian window, so that the correct instantaneous phase is preserved in the Stockwell spectrum (issue #9)
Fix inverse Stockwell transform to match the phase-corrected forward transform
New tests: phase monotonicity and phase step
v1.2 - 2025-01-08
Add support for Python 3.13
Remove support for Python 3.8
v1.1.2 - 2024-05-22
Make it possible again to directly call functions from submodules, like it was in versions prior to v1.1. Example:
import stockwell stockwell.st.st(data, fmin, fmax)
v1.1.1 - 2024-05-21
Remove support for Python 3.6 and 3.7
Add support for Python 3.11 and 3.12
Support for Numpy 2.0!
Added a script to download and compile FFTW3
v1.1 - 2023-06-05
Use ctypes to wrap the modules written in C
License changed to GPLv3
v1.0.7 - 2022-07-27
Packages for Python 3.10 and macOS arm (M1, M2)
v1.0.6 - 2022-04-21
Fix for missing file in source distribution
v1.0.5 - 2022-04-21
New parameters for st.st(): gamma and win_type:
gammacan be used to tune the time and frequency resolutions of the S-transform.win_typecan be set to ‘gauss’ (default) and ‘kazemi’ (Kazemi, 2014)
v1.0.4 - 2021-11-04
Use
versioneerto generate package versionAdd tests
v1.0.3 - 2021-09-15
Automated wheel building and deploy to PyPI
v1.0.2 - 2021-09-15
Updated install instructions in
READMECopy FFTW3 dynamic library into macOS wheel
v1.0.1 - 2021-09-15
Add
numpydependency
v1.0 - 2021-09-15
Initial release