1.0.1

What's Changed

• fixes dead links by @tomjemmett in #149
• 1.0.1 by @yiwen-h in #150

New Contributors

• @tomjemmett made their first contribution in #149

Full Changelog: 1.0.0...1.0.1

1.0.0

What's Changed

• 230831 data by @yiwen-h in #135
• 127 adjust threshold by @yiwen-h in #136
• 137 230908 models by @yiwen-h in #139
• 110 ensembling models by @yiwen-h in #140
• 75 finaltests by @yiwen-h in #143
• 144 v7 framework by @yiwen-h in #145
• 142 remove spacy by @yiwen-h in #146
• 138 final api updates by @yiwen-h in #147
• 1.0.0 by @yiwen-h in #148

Full Changelog: 0.5.5...1.0.0

0.5.5

What's Changed

• Delete build by @ChrisBeeley in #113
• 104 75 linting and tests by @yiwen-h in #114
• added tests for factory_model_performance by @yiwen-h in #116
• 119 datasets by @yiwen-h in #120
• 75 pipeline tests by @yiwen-h in #121
• retrained with multilabel_230719 data by @yiwen-h in #124
• fixed svcpipeline not taking additional_features = False by @yiwen-h in #125
• 117 rulebased model by @yiwen-h in #130
• 126 roc by @yiwen-h in #131
• 118 distilbert api by @yiwen-h in #132
• Security fixes 16 17 25 26 by @yiwen-h in #133
• 0.5.5 by @yiwen-h in #134

Full Changelog: 0.5.4...0.5.5

0.5.4

What's Changed

• 101 two model layers by @yiwen-h in #105
• 90 bert analysis by @yiwen-h in #106
• 107 v6 framework by @yiwen-h in #108
• 107 v6 framework by @yiwen-h in #109
• 97 sentiment api by @yiwen-h in #111
• 0.5.4 by @yiwen-h in #112

Full Changelog: 0.5.3...0.5.4

0.5.3

What's Changed

• minor changes to API to get it working on rsconnect by @yiwen-h in #94
• Api docs by @yiwen-h in #95
• 89 sentiment by @yiwen-h in #96
• 75 moretests by @yiwen-h in #98
• added new dataset and additional_analysis element to pipelines by @yiwen-h in #99
• Security 16 21 by @yiwen-h in #102
• 0.5.3 by @yiwen-h in #103

Full Changelog: 0.5.2...0.5.3

0.5.2

Merge pull request #93 from CDU-data-science-team/0.5.2

updated to 0.5.2 to fix bug 91

0.5.1

Models now trained on subcategories (minor_categories). Narrowed focus to SVC and BERT models, development ceased on other model types.

Further refinements to API to utilise trained SVC model, with improved documentation. API also takes question_type (also referred to as additional_features in the pipeline), which improves performance on specific subcategories.

0.5.0

First multilabel release

Beta version for Connect

Version deployed for demoing to partner trusts

pxtextmining 0.3.4

Fixed typo causing bug & addressed vulnerability issues.