Skip to content

release_notes.rst

Latest commit

 

History

History
285 lines (237 loc) · 10.5 KB

release_notes.rst

File metadata and controls

285 lines (237 loc) · 10.5 KB
Raw

Changelog

v2.13.0 May 15, 2024

  • Changes
    • Temporarily restrict tensorflow version for complete install (:pr:`277`)
    • Remove Mac specific tensorflow requirements (:pr:`281`)
  • Testing Changes
    • Update release notes updated check CI action (:pr:`284`)

Thanks to the following people for contributing to this release: :user:`thehomebrewnerd`

v2.12.0 Feb 26, 2024

Warning

This release of nlp-primitives will not support Python 3.8

  • Changes
    • Remove support for Python 3.8 and add support for 3.11 (:pr:`269`)
  • Testing Changes
    • Update release.yaml to use trusted publisher for PyPI releases (:pr:`269`)

Thanks to the following people for contributing to this release: :user:`thehomebrewnerd`

v2.11.0 Apr 13, 2023

  • Fixes
    • Fix Makefile package command (:pr:`241`)
  • Changes
    • Fix ruff argument in pyproject.toml (:pr:`239`)
    • Remove constants.py (:pr:`243`)
    • Remove Woodwork as a core requirement (:pr:`258`)
  • Testing Changes
    • Add pull request check for linked issues to CI workflow (:pr:`245`)

Thanks to the following people for contributing to this release: :user:`gsheni`, :user:`sbadithe`

v2.10.0 Jan 10, 2023

  • Enhancements
    • Add conda create feedstock pull request workflow (:pr:`220`)
    • Improve PartOfSpeech docstring (:pr:`224`)
  • Fixes
    • Fix handling of all-whitepace strings in MeanCharactersPerSentence (:pr:`234`)
  • Changes
    • Update workflow_dispatch to release workflow (:pr:`221`)
    • Add ruff, remove isort, and add more pre-commits (:pr:`237`)
  • Testing Changes
    • Add pytest-xdist as test requirement and use auto option to use all cores when running unit tests (:pr:`218`)
    • Adds Windows install test (:pr:`219`)

Thanks to the following people for contributing to this release: :user:`gsheni`, :user:`sbadithe`

v2.9.0 Oct 24, 2022

  • Fixes
    • Fixes erroneous case-sensitive stopword checking in clean_tokens (:pr:`211`)
  • Changes
    • Remove primitives that were transferred to Featuretools (:pr:`214`)

Thanks to the following people for contributing to this release: :user:`sbadithe`, :user:`thehomebrewnerd`

Breaking Changes

  • Multiple primitives were removed from nlp_primitives and transferred to Featuretools as standad primitives. See :pr:`214` for a

complete list of primitives that were moved. These primitives can now be imported directly from Featuretools. For example, use from featuretools.primitives import CountString instead of the previous import of from nlp_primitives import CountString.

v2.8.0 Sep 14, 2022

  • Enhancements
    • Add NumberOfHashtags and NumberOfMentions` primitives (:pr:`180`)
    • Add NumberOfUniqueWords primitive (:pr:`187`)
    • Add NumberOfSentences and MeanSentenceLength primitives (:pr:`188`)
    • Add NumberOfWordsInQuotes primitive (:pr:`196`)
  • Fixes
    • Update README.md with Alteryx info (:pr:`167`)
  • Changes
    • Add Woodwork as core dependency (:pr:`170`)
    • Add support for Python 3.10 (:pr:`175`)
    • Drop support for Python 3.7 (:pr:`176`)
    • Change TitleWordCount, PunctuationCount, UpperCaseCount to use CountString (:pr:`183`)
    • Remove readthedocs and docs requirements (:pr:`193`)
    • Use pyproject.toml only (move away from setup.cfg) (:pr:`201`)
  • Testing Changes
    • Change codecov v3 for GitHub workflow (:pr:`184`)

Thanks to the following people for contributing to this release: :user:`gsheni`, :user:`sbadithe`, :user:`thehomebrewnerd`

v2.7.1 Jun 29, 2022

  • Fixes
    • Clean up naming of LSA features to prevent full custom corpus from being displayed (:pr:`161`)

Thanks to the following people for contributing to this release: :user:`thehomebrewnerd`

v2.7.0 Jun 16, 2022

  • Enhancements
    • Allow users to optionally pass in a custom corpus to use with the LSA primitive (:pr:`148`)
  • Fixes
    • Fix bug in CountString with null values (:pr:`154`)
    • Fix a bug with nltk data was not included in package (:pr:`157`)
  • Documentation Changes
    • Update release branch naming convention in documentation (:pr:`155`)
  • Testing Changes
    • Add workflow to test nlp_primitives without test dependencies (:pr:`157`)

Thanks to the following people for contributing to this release: :user:`gsheni`, :user:`rwedge`, :user:`thehomebrewnerd`

v2.6.0 Jun 16, 2022

  • Changes
    • Transition to use pyproject.toml and setup.cfg (moving away from setup.py) (:pr:`127`, :pr:`132`)
    • Elmo and UniversalSentenceEncoder added to the nlp_primitives.tensorflow module namespace (:pr:`150`)
  • Testing Changes
    • Fix latest dependency checker to create PR (:pr:`129`)
    • Fixed unit tests workflow test choice logic (:pr:`151`)

Thanks to the following people for contributing to this release: :user:`gsheni`, :user:`rwedge`, :user:`thehomebrewnerd`

v2.5.0 Apr 7, 2022

  • Fixes
    • Fix NumUniqueSeparators to allow for serialization and deserialization (:pr:`122`)
  • Changes
    • Speed up LSA primitive initialization (:pr:`118`)
  • Testing Changes
    • Fix install test and update Makefile (:pr:`123`)

Thanks to the following people for contributing to this release: :user:`rwedge`, :user:`thehomebrewnerd`

v2.4.0 Mar 31, 2022

  • Changes
    • Added pip dependencies for M1 Macs (:pr:`117`)
  • Testing Changes
    • Added Release Notes CI Check (:pr:`110`)
    • Added CI check to ensure entrypoint works with Featuretools (:pr:`111`)
    • Fixed workflow that tests latest changes to featuretools (:pr:`112`)

Thanks to the following people for contributing to this release: :user:`dvreed77`, :user:`gsheni`, :user:`jeff-hernandez`, :user:`thehomebrewnerd`

v2.3.0 Feb 28, 2022

  • Changes
    • Tensorflow dependent primitives only imported at top level if tensorflow is installed (:pr:`105`)
  • Testing Changes
    • Skip Tensorflow dependent tests if --notensorflow flag is passed to pytest (:pr:`107`)

Thanks to the following people for contributing to this release: :user:`dvreed77`, :user:`rwedge`

v2.2.0 Feb 17, 2022

  • Enhancements
    • Add NumberOfUniqueSeparators primitive (:pr:`90`)
    • Add NumberOfCommonWords primitive (:pr:`92`)
    • Add CountString and WhitespaceCount primitives (:pr:`89`)
    • Add entry point for featuretools primitives (:pr:`98`)
  • Fixes
    • Fixes error with MeanCharactersPerWord primitive with series of Nones (:pr:`101`)
  • Documentation Changes
    • Remove testing on conda forge in release.md (:pr:`84`)
  • Testing Changes
    • Update scheduler workflow to use correct name and file (:pr:`87`, :pr:`86`)
    • Add workflow to auto-merge dependency PRs (:pr:`93`)

Thanks to the following people for contributing to this release: :user:`dvreed77`, :user:`gsheni`, :user:`jeff-hernandez`, :user:`tuethan1999`

v2.1.0 Dec 21, 2021

  • Enhancements
    • Add primitive for total word length TotalWordLength (:pr:`79`)
    • Add primitive for median word length MedianWordLength (:pr:`80`)
  • Changes
    • Update setup.py with new and correct information (:pr:`72`)
  • Testing Changes
    • Add python 3.9 CI for unit tests and entrypoint tests (:pr:`72`)

Thanks to the following people for contributing to this release: :user:`gsheni`, :user:`jeff-hernandez`

v2.0.0 Oct 13, 2021

  • Changes
    • Update primitives for compatibility with Featuretools 1.0.0 (:pr:`61`)
  • Testing Changes
    • Individual CI jobs will not cancel if other jobs fail (:pr:`67`)

Thanks to the following people for contributing to this release: :user:`rwedge`, :user:`thehomebrewnerd`

v1.2.0 Sept 3, 2021

  • Enhancements
    • Add Elmo primitive (:pr:`64`)
  • Changes
    • Drop python 3.6 support (:pr:`57`)
  • Documentation Changes
    • Update UniversalSentenceEncoder docstring example (:pr:`42`)

Thanks to the following people for contributing to this release: :user:`davesque`, :user:`gsheni`, :user:`jeff-hernandez`, :user:`rwedge`

Breaking Changes

  • Drop python 3.6 support (:pr:`57`)

v1.1.0 Oct 26, 2020

  • Changes
    • Upgrade to Featuretools 0.20.0 and replace Text variable type with NaturalLanguage (:pr:`34`)
    • Include corpus download with package instead of downloading at first run. (:pr:`31`)
    • Change corpus used by LSA primitive (:pr:`35`)
    • Refactor clean_tokens function (:pr:`29`)
  • Testing Changes
    • Add another set of CI jobs which ensure that we can make an nlp_primitives package, install that, and then still pass all the unit tests, vs an editable install :pr:`31`
    • Move some test utils into test/ :pr:`31`

Thanks to the following people for contributing to this release: :user:`dsherry`, :user:`eccabay`, :user:`gsheni`, :user:`rwedge`

v1.0.0 Aug 12, 2020

  • Changes
    • Remove tensorflow and tensorhub as core requirements, but they

    can be installed with pip install nlp_primitives[complete]. The UniversalSentenceEncoder primitive requires the nlp_primitives[complete] install but all other primitives work with the standard install. (:pr:`24`)

  • Testing Changes
    • Update CircleCI to perform complete install and use matrix jobs (:pr:`24`)

Thanks to the following people for contributing to this release: :user:`thehomebrewnerd`

v0.3.1

  • Fix installation error related to scipy version

v0.3.0

  • Fixed case-insensitivity in the Stopword Count Primitive
  • Made compatible with Tensorflow 2
  • Dropped Python 3.5 and added Python 3.8

v0.2.5

  • Removed python-dateutil as a requirement

v0.2.4

  • Added Featuretools Entry Point
  • PyPI Upload

v0.2.3

  • Small bug fixes

v0.2.2

  • Now comes with description for PyPI

v0.1.0

  • Fixed reliance on external data files

v0.0.0

  • Initial Release