flair-uk-ner
Model description
flair-uk-ner is a Flair model that is ready to use for Named Entity Recognition. It is based on flair embeddings, that I've trained for Ukrainian language (available here and here) and has nice performance and a very small size (just 72mb!).
It has been trained to recognize four types of entities: location (LOC), organizations (ORG), person (PERS) and Miscellaneous (MISC).
Results:
- F-score (micro) 0.8605
- F-score (macro) 0.7472
- Accuracy 0.8033
by class | precision | recall | f1-score | support |
---|---|---|---|---|
PERS | 0.9305 | 0.9422 | 0.9363 | 1678 |
LOC | 0.8150 | 0.8678 | 0.8406 | 401 |
ORG | 0.6653 | 0.6092 | 0.6360 | 261 |
MISC | 0.6202 | 0.5375 | 0.5759 | 240 |
micro avg | 0.8616 | 0.8593 | 0.8605 | 2580 |
macro avg | 0.7577 | 0.7392 | 0.7472 | 2580 |
weighted avg | 0.8569 | 0.8593 | 0.8575 | 2580 |
The model was fine-tuned on the NER-UK dataset, released by the lang-uk. Training code is also available here.
Copyright: Dmytro Chaplynskyi, lang-uk project, 2022
- Downloads last month
- 12
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Evaluation results
- NER Precisionself-reported0.862
- NER Recallself-reported0.859
- NER F Scoreself-reported0.861