nlp-uoregon/trankit
A multilingual NLP toolkit using transformer models to perform tokenization, parsing, and tagging across 56 languages.
Not currently ranked — collecting fresh signals.
star history
Trankit is a Python toolkit for multilingual natural language processing built on PyTorch and transformer architectures. It provides pre-trained pipelines for tasks including sentence segmentation, tokenization, part-of-speech tagging, morphological tagging, lemmatization, and dependency parsing. The toolkit supports 56 languages using XLM-RoBERTa-based models and offers both command-line and Python API interfaces.
Frequently asked
- What is nlp-uoregon/trankit?
- A multilingual NLP toolkit using transformer models to perform tokenization, parsing, and tagging across 56 languages.
- Is trankit open source?
- Yes — nlp-uoregon/trankit is open source, released under the Apache-2.0 license.
- What language is trankit written in?
- nlp-uoregon/trankit is primarily written in Python.
- How popular is trankit?
- nlp-uoregon/trankit has 795 stars on GitHub.
- Where can I find trankit?
- nlp-uoregon/trankit is on GitHub at https://github.com/nlp-uoregon/trankit.