v4.1.2

What's Changed

• Handle more than 8 compressed signals in wrsamp by @bemoody in #450
• Add a test to check if wrsamp can write more than 8 channels to format 516 by @alistairewj in #451
• Use int64 instead of int for ann2rr by @bemoody in #454

Full Changelog: v4.1.1...v4.1.2

v4.1.1

• Removing upper bounds on dependencies (#448, #447)
• Miscellaneous style and typing fixes (#445, #440)

v4.1.0

• Converting a record into a Pandas DataFrame (#380)
• Locating signals in a multi-segment record (#403)
• Writing custom annotation symbols (#404)
• Performance improvements when writing annotations (#406)
• Correct rounding when converting floating-point signal data (#419)
• Writing signals in compressed format (#420)
• Decoding non-ASCII text in EDF files (#429)
• Bug fixes when writing signal metadata (#424)
• Documentation updates (#401, #407, #433, #437)
• Internal refactoring (#396, #400)
• Test suite improvements (#416, #422)

v4.0.0

• Supports reading WFDB records that contain FLAC signal files. #372
• Switches to Poetry for package management. #356
• Sets Python support to >=3.7
• Supports reading multi-frequency multi-segment records. #331
• Supports plotting multi-frequency data using plot_wfdb. ​​#348
• Plots annotations on top of signals in plot functions. #345
• Aligns signals horizontally in plot_wfdb if possible by default. #390
• Correctly handles sample numbers > 2**31 in annotation files. #328
• Fixes bugs when reading formats 8, 310, and 311. #327
• Removes the ability for rdrecord and rdsamp to read EDF and WAV files. #370
• Renames edf2mit, mit2edf, wav2mit, mit2wav, and csv2mit to read_edf, wfdb_to_edf, read_wav, wfdb_to_wav, and csv_to_wfdb respectively. #370 See also this note.
• Moves the above format conversion functions to the wfdb.io.convert submodule. #370

v3.4.1

This release makes the following changes:

• Fix axes indexing issue when plotting a single channel: bugfix: unable to plot single channel record #308
• Update matplotlib version and import only when used: This updates the Matplotlib version to avoid segmentation faults on Macs running Big Sur. #310 and Lazily import matplotlib #314
• Performance improvements to smooth_frames=False: Optimize smooth_frames=False #318
• List only direct dependencies in requirements.txt: Remove transitive dependencies #322
• Correct an error in the calculation of the signals length for certain cases: Correctly infer length of record when unspecified #323
• Speed up import time by removing the dependency on scikit-learn when it isn't required: Remove dependency on scikit-learn #321

v3.4.0

• Adds EDF+C support for edf2mit
• Improves speed of resample_ann
• Allows negative annotation subtypes
• Adds rr2ann
• Updates to Python 3.6+
• Adds sharex and sharey to plot_items
• Adds wfdbdesc
• Adds wfdbtime
• Adds conversion for CSV to WFDB-annotations
• Adds rdedfann
• Adds information about WFDB header format in wfdb/io/_header.py
• Adds mrgann
• Adds sigavg

v3.3.0

• Adds CSV to WFDB file converter (also TSV, and any other valid delimiter)
• Updates plotting labels functionality
• Updates documentation to include more functions

v3.2.0

• Removes the MNE package in replacement of direct binary file reading for EDF files for increased speed
• EDF file reading bug fixes
• Relaxing of requirements for pytz

3.1.2

• Fixes some problems on Windows machines
• Updates Pandas syntax for most recent version
• Cleans some package requirements
• Adds more efficient EDF header reader

v3.1.1

• Fixes error when downsampling single channel if NaN values are present.
• Fixes URL generation for Windows machines by removing the os.sep operation.