v0.31.0

v0.31.0 May 13, 2024

• Enhancements
  • Add support for Python 3.12 #1855
• Changes
  • Drop support for using Woodwork with Dask or Pyspark dataframes #1857
  • Use filter arg in call to tarfile.extractall to safely deserialize DataFrames #1862

Thanks to the following people for contributing to this release:
@thehomebrewnerd

v0.30.0

v0.30.0 Apr 10, 2024

Support for use with Dask and Pyspark dataframes is planned for removal in an upcoming release of Woodwork.

• Changes
  • Temporarily restrict Dask version #1837
  • Updates for compatibility with Dask 2024.4.1 #1843
• Testing Changes
  • Fix serialization test to work with pytest 8.1.1 #1837

Thanks to the following people for contributing to this release:
@thehomebrewnerd

v0.29.0

v0.29.0 Feb 26, 2024

• Changes
  • Remove numpy upper bound restriction in pyproject.toml #1819
  • Bump min version of python-dateutil for pandas 2.0 compatibility #1825
• Testing Changes
  • Update release.yaml to use trusted publisher for PyPI releases #1819
  • Update latest dependency CI runs to include run with only core requirements #1822

Thanks to the following people for contributing to this release:
@thehomebrewnerd

v0.28.0

Warning: This release of Woodwork will not support Python 3.8

• Changes
  • Upgraded numpy to < 2.0.0 #1799
• Documentation Changes
  • Added dask string storage note to "Other Limitations" in Dask documentation #1799
• Testing Changes
  • Upgraded moto and boto3 #1799

@cp2boston, @gsheni, @tamargrey

v0.27.0

v0.27.0 Dec 12, 2023

• Fixes
  • Removed warning due to deprecated infer_datetime_format argument in pandas (#1785)
  • Fix GitHub Actions to kick off EvalML and Featuretools unit tests (#1795)
• Changes
  • Temporarily restrict pyarrow version due to serialization issues (#1768)
  • Update pandas categorical type call and remove black with ruff formatter (#1794)
• Testing Changes
  • Removed old performance testing workflow (#1776)

Thanks to the following people for contributing to this release:
@eccabay, @gsheni, @thehomebrewnerd, @petejanuszewski1

v0.26.0

v0.26.0 August 22, 2023

• Enhancements
  • Optimized Boolean inference by removing generation of mappings and sets of boolean values (#1713)
  • Speed up Boolean and Integer inference by caching results of corresponding nullable type inference (#1733)
• Fixes
  • Update s3 bucket for docs image (#1749)
• Documentation Changes
  • Update readthedocs config to use build.os (#1753)
  • Fix PyPI badge not showing on README.md (#1755)

Thanks to the following people for contributing to this release:
@gsheni, @sbadithe, @simha104

v0.25.1

v0.25.1 Jul 18, 2023

• Fixes
  • Restrict numpy version to resolve boolean inference issue with v1.25.0 #1735

Thanks to the following people for contributing to this release:
@thehomebrewnerd

v0.25.0

v0.25.0 Jul 17, 2023

• Enhancements
  • Force datetime guesser input to be string #1724
  • Add support for pandas v2.0.0 #1729
• Changes
  • Remove upper bound restriction on dask version #1729
• Testing Changes
  • Remove autouse=True from latlong dataframe fixtures #1729

Thanks to the following people for contributing to this release:
@christopherbunn, @thehomebrewnerd

v0.24.0

v0.24.0 May 24, 2023

• Enhancements
  • Removed repeated sorting for numeric data in _get_describe_dict to improve performance (#1682)
  • Improved inference for URL, EmailAddress, and other logical types by defining new parent-child relationships (#1702)
  • Added an include_time_index argument when calculating dependence measures (#1698)
• Changes
  • Stopped calculating top_values for Double columns with integer values (#1692)
• Testing Changes
  • Add Python 3.11 markers, add 3.11 for unit tests & install test (#1678)
  • Run looking glass performance tests on merge via Airflow (#1695)

Thanks to the following people for contributing to this release:
@bchen1116, @gsheni, @ParthivNaresh, @petejanuszewski1, @simha104, @tamargrey

v0.23.0

v0.23.0 April 12, 2023

• Fixes
  • Updated Datetime format inference to include formats with two digit year dates along with timezones (#1666)
• Changes
  • Updated add_type and remove_type to include a treatment argument (#1661)
  • Limit pandas <2.0.0 for core requirements (#1668)
  • Upgrade minimum dask to 2022.11.1 and minimum pandas to 1.4.3 (#1671)

Thanks to the following people for contributing to this release:
@gsheni, @jeff-hernandez, @ParthivNaresh, @simha104