0.05

What's Changed

• Fix handling of dataframe input, issue #300 by @toni-neurosc in #301
• Move tox.ini contents to pyproject.toml by @toni-neurosc in #308
• Improve Github actions testing workflow by @toni-neurosc in #311
• Fix type hints, clean imports, and other errors reported by static analysis by @toni-neurosc in #305
• Add NMPreprocessor protocol and Preprocessors class by @toni-neurosc in #317
• Turn Feature abstract class into a Protocol and lazy load feature modules by @toni-neurosc in #307
• Add logging class by @toni-neurosc in #306
• Import all modules as lazily as possible by @toni-neurosc in #319
• Add mne lsl feature by @SamedVossberg in #323
• Refactor lsl stream by @SamedVossberg in #326
• Move LSL lib files inside module directory, increase OS support, fix docs build by @toni-neurosc in #328
• Generate results directly from feature_dict instead of pd.Series by @toni-neurosc in #334
• Pydantic validation settings by @timonmerk in #329
• Timing lsl by @SamedVossberg in #337
• MNEFilter with fftconvolve by @toni-neurosc in #339
• Optimize Hjorth and LineLength calculation by @toni-neurosc in #340
• Add compatibility with Numpy 1.26 and 2.0 by @toni-neurosc in #341
• Fix docstrings by @timonmerk in #343
• MNE connectivity 2x speed-up by @toni-neurosc in #346
• Remove features dictionary as input to calc_feature by @toni-neurosc in #349
• Nm database by @SamedVossberg in #350

New Contributors

• @SamedVossberg made their first contribution in #323

Full Changelog: v0.03...0.05

Bugfix build dependencies

What's Changed

• Fix handling of dataframe input, issue #300 by @toni-neurosc in #301

Full Changelog: v0.03...0.0.4

Improved performances

What's Changed

• add Welch method by @timonmerk in #269
• Add plotter raw by @timonmerk in #270
• fix ch_used data selection by @timonmerk in #274
• rename py_neuromodulation consistent to nm by @timonmerk in #275
• add option to store psd for fft, stft and welch method by @timonmerk in #276
• Fix add targets by @timonmerk in #284
• Fix broken link by @toni-neurosc in #287
• Save CSVs with PyArrow by @toni-neurosc in #280
• Parallel processing pr by @timonmerk in #290
• Avoid use of Numpy nan statistics if possible by @toni-neurosc in #285
• Bursts feature performance improvement by @toni-neurosc in #286
• Sharpwaves performance fixes by @toni-neurosc in #288
• Add preprocessing filter by @timonmerk in #294
• Logging by @timonmerk in #298
• Add R-Map computer without Lead-Mapper by @timonmerk in #299

New Contributors

• @toni-neurosc made their first contribution in #287

Full Changelog: v0.02...v0.03

Added bispectrum and fix burst computation

• #267 Thanks to @tsbinns added feature of bispectrum https://github.com/braindatalab/PyBispectra
• #266 Fixed bug in nm_bursts, s.t. s["bandpass_filter_settings"]["segment_lengths_ms"] restricts threshold computation
• add np.float64 cast s.t. mne notch_filter does not fail with int types

Preprint version

Preprint: https://doi.org/10.21203/rs.3.rs-3212709/v1
Documentation: https://py-neuromodulation.readthedocs.io/en/latest/index.html