Trankit is a light-weight Transformer-based Toolkit for multilingual Natural Language Processing (NLP). It
provides a trainable pipeline for fundamental NLP tasks over 100 languages, and 90 pretrained pipelines for 56
languages.
Trankit can be easily installed via pip: pip install trankit
For more information, please check out our github repo ,
documentation , and technical paper .
Usage
Text to annotate
Afrikaans
Ancient Greek (Default)
Ancient Greek (Perseus)
Arabic
Armenian
Basque
Belarusian
Bulgarian
Catalan
Chinese (Classical)
Chinese (Simplified)
Chinese (Traditional)
Croatian
Czech (CAC)
Czech (CLTT)
Czech (Default)
Czech (FicTree)
Danish
Dutch (Default)
Dutch (LassySmall)
English (Default)
English (GUM)
English (LinES)
English (ParTUT)
Estonian (Default)
Estonian (EWT)
Finnish (Default)
Finnish (FTB)
French (Default)
French (ParTUT)
French (Sequoia)
French (Spoken)
Galician (Default)
Galician (TreeGal)
German (Default)
German (HDT)
Greek
Hebrew
Hindi
Hungarian
Indonesian
Irish
Italian (Default)
Italian (ParTUT)
Italian (PoSTWITA)
Italian (TWITTIRO)
Italian (VIT)
Japanese
Kazakh
Korean (Default)
Korean (Kaist)
Kurmanji
Latin (Default)
Latin (PROIEL)
Latin (Perseus)
Latvian
Lithuanian (Default)
Lithuanian (HSE)
Marathi
Norwegian (Bokmaal)
Norwegian (Nynorsk)
Norwegian (NynorskLIA)
Old French
Old Russian
Persian
Polish (Default)
Polish (LFG)
Portuguese (Default)
Portuguese (GSD)
Romanian (Default)
Romanian (Nonstandard)
Russian (Default)
Russian (GSD)
Russian (Taiga)
Scottish Gaelic
Serbian
Slovak
Slovenian (Default)
Slovenian (SST)
Spanish (Default)
Spanish (GSD)
Swedish (Default)
Swedish (LinES)
Tamil
Telugu
Turkish
Ukrainian
Urdu
Uyghur
Vietnamese
Annotate
Visualization: Brat Rapid Annotation Tool.