Sefef

Latest version: v3.0.0

Safety actively analyzes 723685 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 3 of 5

2.1.0

What's Changed
* TSCV now removes (defaults to True) non-preictal and non-interictal samples from train set.

2.0.3

What's Changed
* Minor changes: postprocessing no longer removes last forecast samples in case of incomplete window of forecast

2.0.1

What's Changed
* [minor correction: condition in tscv criteria check]

2.0.0

What's Changed
* TSCV now takes into consideration missing preictal data before seizure onsets
* Removed unused variable in _evaluation.Dataset_

Attribute errors will be raised in _evaluation_ module with previous code versions

1.5.0

What's Changed
* Visualization of forecasts now follows the shape of a step function (with duration equal to forecast horizon), instead of point estimate at the start of the forecast horizon.
* Also realized that title of plot was always "Daily event likelihood" and it can now be a user-defined parameter.

1.4.0

What's Changed
* Added possibility of returning train and test data from specified fold

**Full Changelog**: https://github.com/anascacais/SeFEF/compare/v1.0.0...v1.4.0

Page 3 of 5

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.