dchaplinsky
commited on
Commit
•
11b3a94
1
Parent(s):
2ade44e
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,58 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
license: mit
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
tags:
|
3 |
+
- flair
|
4 |
+
- token-classification
|
5 |
+
- sequence-tagger-model
|
6 |
+
language: uk
|
7 |
+
datasets:
|
8 |
+
- ner-uk
|
9 |
+
model-index:
|
10 |
+
- name: flair-uk-ner
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: NER
|
14 |
+
type: token-classification
|
15 |
+
metrics:
|
16 |
+
- name: NER Precision
|
17 |
+
type: precision
|
18 |
+
value: 0.8572
|
19 |
+
- name: NER Recall
|
20 |
+
type: recall
|
21 |
+
value: 0.8516
|
22 |
+
- name: NER F Score
|
23 |
+
type: f_score
|
24 |
+
value: 0.8544
|
25 |
+
widget:
|
26 |
+
- text: "Президент Володимир Зеленський пояснив, що наразі діалог із режимом Володимира путіна неможливий, адже агресор обрав курс на знищення українського народу. За словами Зеленського цей режим РФ виявляє неповагу до суверенітету і територіальної цілісності України."
|
27 |
license: mit
|
28 |
---
|
29 |
+
|
30 |
+
# flair-uk-ner
|
31 |
+
|
32 |
+
## Model description
|
33 |
+
|
34 |
+
**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](https://huggingface.co/dchaplinsky/flair-uk-backward) and [here](https://huggingface.co/dchaplinsky/flair-uk-forward)) and has nice performance and a very **small size** (just 77mb!).
|
35 |
+
|
36 |
+
It has been trained to recognize four types of entities: location (LOC), organizations (ORG), person (PERS) and Miscellaneous (MISC).
|
37 |
+
|
38 |
+
Results:
|
39 |
+
- F-score (micro) 0.8544
|
40 |
+
- F-score (macro) 0.7406
|
41 |
+
- Accuracy 0.798
|
42 |
+
|
43 |
+
precision recall f1-score support
|
44 |
+
|
45 |
+
PERS 0.9231 0.9374 0.9302 1678
|
46 |
+
LOC 0.8204 0.8429 0.8315 401
|
47 |
+
ORG 0.6708 0.6245 0.6468 261
|
48 |
+
MISC 0.6029 0.5125 0.5541 240
|
49 |
+
|
50 |
+
micro avg 0.8572 0.8516 0.8544 2580
|
51 |
+
macro avg 0.7543 0.7293 0.7406 2580
|
52 |
+
weighted avg 0.8518 0.8516 0.8512 2580
|
53 |
+
|
54 |
+
The model was fine-tuned on the [NER-UK dataset](https://github.com/lang-uk/ner-uk), released by the [lang-uk](https://lang.org.ua).
|
55 |
+
Training code is also available [here](https://github.com/lang-uk/flair-ner).
|
56 |
+
|
57 |
+
|
58 |
+
Copyright: Dmytro Chaplynskyi, [lang-uk project](https://lang.org.ua), 2022
|