Instructions to use ukr-models/uk_core_news_trf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- spaCy
How to use ukr-models/uk_core_news_trf with spaCy:
!pip install https://huggingface.co/ukr-models/uk_core_news_trf/resolve/main/uk_core_news_trf-any-py3-none-any.whl # Using spacy.load(). import spacy nlp = spacy.load("uk_core_news_trf") # Importing as module. import uk_core_news_trf nlp = uk_core_news_trf.load() - Notebooks
- Google Colab
- Kaggle
Spacy transformer pipeline for Ukrainian language (XLM-Roberta based). Components: transformer, ner, morphologizer, parser. Training details
- Downloads last month
- 7
Evaluation results
- NER Precisionself-reported0.889
- NER Recallself-reported0.890
- NER F Scoreself-reported0.889
- POS (UPOS) Accuracyself-reported0.983
- Morph (UFeats) Accuracyself-reported0.961
- Unlabeled Attachment Score (UAS)self-reported0.962
- Labeled Attachment Score (LAS)self-reported0.946
- Sentences F-Scoreself-reported0.930