v1.2.0

New Features

This release adds the following features to Openseize:

1. Producers now support concurrent processing. This feature supports applying a single DSP pipeline across multiple producer instances simultaneously. To use this feature, a sequence of DSP steps should be constructed as a single function to pass to Python's multiprocessing module like so:

# call an openseize pipeline on a sequence of producers
 with mp.Pool(processes=ncores) as pool:
            results = []
            for res in pool.imap(pipeline, producers):
                results.append(res)

Please see the test_concurrency pytest module for complete examples and tests.

2. The shape attribute of EDF Reader instances have been promoted to an attribute of the Reader ABC. This ensures that all future readers implement a shape method. The documentation and tutorials have been updated to reflect this change.

3. A new tutorial details how to use Openseize with non-EDF files. In particular, a worked example is provided to developers seeking to write new Reader types.

Compatibility

There are no compatibility issues with the prior v1.1.0 release.

Bug Fixes

A minor documentation fix was applied to the docstring of the IIR notch

v1.1.0-joss

This is the stable release of Openseize as described in Openseize: A digital signal processing package for large EEG datasets in Python JOSS 2023. The DOI of this release issued by Zenodo is 10.5281/zenodo.7760376

v1.1.0

New Features

This release adds the following features to Openseize:

1. openseize.file_io.path_utils: Module containing utilities for working with path instances including:

   • regex matching of file stems of different types
   • regex extraction of metadata stored in file stems
   • in-place renaming of files using regex search and replace

2. openseize.file_io.dialogs.matching: A modal dialog for regex matching of files of different types either by ui selection of files, ui selection of a parent directory or an argument supplied directory.

Compatibility

The openseize.file_io.path_utils replaces the openseize.tools.paths module which has been removed.

Bug Fixes

Spelling errors throughout the source code and documentation have been corrected with codespell. This dependency has been added to the develop.yml environment and the pyproject.toml files.

v1.0.0

New Features

This is the initial stable release of Openseize, a Python package for digital signal processing of big EEG datasets.

Compatibility

A compatibility report will be available on future releases.

Bug Fixes

A bug fix report will be available on future releases.