V3.3.0 on PyPi

• updated build (for PEP 621, etc.)
• using nx.DiGraph instead
• new release procedures

What's Changed

• update scrubber documentation by @Ankush-Chander in #238
• [Snyk] Security upgrade tornado from 6.2 to 6.3.3 by @ceteri in #240
• [Snyk] Security upgrade tornado from 6.2 to 6.3.3 by @ceteri in #250
• make lemma_graph undirected by @Ankush-Chander in #264
• [Snyk] Security upgrade jupyter-server from 1.24.0 to 2.11.2 by @ceteri in #270
• [Snyk] Security upgrade werkzeug from 2.2.3 to 3.0.1 by @ceteri in #268
• build updates by @ceteri in #273

Full Changelog: v3.2.5...v3.3.0

v3.2.5 release on PyPi

2023-08-07

• improved scrubber; kudos @0dB @Ankush-Chander
• updated dependencies, responding to vuln scanner

What's Changed

• updating the GH actions to v3 by @ceteri in #235
• Scrubber sample code now handles "singular" tokens correctly by @0dB in #233
• prep release 3.2.5 by @ceteri in #237

New Contributors

• @0dB made their first contribution in #233

Full Changelog: v3.2.4...v3.2.5

v3.2.4 release on PyPi

2022-07-27

• better support for "ru" and other languages without noun_chunks support in spaCy
• updated example notebook to illustrate TopicRank algorithm
• made the node bias setting case-independent for Biased Textrank algorithm; kudos @Ankush-Chander
• updated summarization tests; kudos @tomaarsen
• reworked some unit tests to be less brittle, less dependent on specific spaCy point releases

What's Changed

• updated docs and example to show TopicRank by @ceteri in #211
• working on #204 by @ceteri in #212
• Prevent exception on TopicRank when there are no noun_chunks by @tomaarsen in #219
• Biasedrank case fix by @Ankush-Chander in #217
• Docs update by @ceteri in #221
• rework some unit tests by @ceteri in #222

Full Changelog: v3.2.3...v3.2.4

v3.2.3 release on PyPi

2022-03-06

• handles missing noun_chunks in some language models (e.g., "ru") #204
• add TopicRank algorithm; kudos @tomaarsen
• improved test suite; fixed tests for newer spacy releases; kudos @tomaarsen

What's Changed

• [Snyk] Security upgrade mistune from 0.8.4 to 2.0.1 by @snyk-bot in #201
• Improved test suite; fixed tests by @tomaarsen in #205
• Updated Copyright year from 2021 to 2022 by @tomaarsen in #206
• update API reference docs by @ceteri in #207
• Inclusion of the TopicRank Keyphrase Extraction algorithm by @tomaarsen in #208
• Prep release by @ceteri in #210

New Contributors

• @snyk-bot made their first contribution in #201

Full Changelog: v3.2.2...v3.2.3

v3.2.2 release on PyPi

What's Changed

• prep next release by @ceteri in #189
• warning about the deprecated code in archive by @ceteri in #190
• fixes chunk to be between sent_start and sent_end in BaseTextRank.calc_sent_dist by @clabornd in #191
• Update by @ceteri in #198
• add more scrubber examples and documentation by @dayalstrub-cma in #197
• kudos by @ceteri in #199
• prep PyPi release by @ceteri in #200

New Contributors

• @clabornd made their first contribution in #191
• @dayalstrub-cma made their first contribution in #197

Full Changelog: v3.2.1...v3.2.2

v3.2.1 release on PyPi

2021-07-24

• add "paragraph" option into summary() function; kudos @CaptXiong

v3.2.0 release on PyPi

2021-07-17

Various support for spaCy 3.1.x updates, which changes some interfaces.

• NB: THE SCRUBBER UPDATE WILL BREAK PREVIOUS RELEASES
• allow Span as scrubber argument, to align with spaCy 3.1.x; kudos @Ankush-Chander
• add lgtm code reviews (slow, not integrating into GitHub PRs directly)
• evaluating grayskull to generate a conda-forge recipe
• add use of pipdeptree to analyze dependencies
• use KG from biblio.ttl to generate bibliography
• fixed overlooked comment from earlier code; kudos @debraj135
• add visualisation using altair; kudos @louisguitton
• add scrubber usage in sample notebook; kudos @Ankush-Chander
• integrating use of MkRefs to generate semantic reference pages in docs

v3.1.1 release on PyPi

2021-03-25

• fix the span length calculation in explanation notebook; kudos @Ankush-Chander
• add BiasedTextRank by @Ankush-Chander (many thanks!)
• add conda environment.yml plus instructions
• use bandit to check for security issues
• use codespell to check for spelling errors
• add pre-commit checks in general
• update doc._.phrases in the call to change_focus() so the summarization will sync with the latest focus

v3.1.0 release on PyPi

2021-03-12

• rename master branch to main
• add a factory class that assigns each doc its own Textrank object; kudos @Ankush-Chander
• refactor the stopwords feature as a constructor argument
• add get_unit_vector() method to expose the characteristic unit vector
• add calc_sent_dist() method to expose the sentence distance measures (for summarization)
• include a unit test for summarization
• updated contributor instructions
• pylint coverage for code checking
• linking definitions and citations in source code apidocs to our online docs
• updated links on PyPi

v3.0.1 release on PyPi

2021-02-27

• mypy coverage for type annotations
• add DOI to README and CITATION
• now deploying online docs at https://derwen.ai/docs/ptr/