Hugging Face's logo

risqaliyevds
/
xlm-roberta-large-ner

Token Classification
Transformers
Safetensors
Uzbek
xlm-roberta
ner
uzbek_ner
ner_for_uzbek_language
Inference Endpoints
Model card Files Community
xlm-roberta-large-ner
File size: 3,320 Bytes 
1b4ce9d
 
 
 
 
 
 
 
 
 
 
 
276e493
1b4ce9d
 
f077131
1b4ce9d
276e493
 
 
1b4ce9d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5706d8e
 
 
 
1b4ce9d
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
 
---
license: mit
language:
- uz
metrics:
- accuracy
pipeline_tag: token-classification
tags:
- ner
- uzbek_ner
- ner_for_uzbek_language
---
## O'zbek tili uchun Named Entity Recognition (NER) modeli

### Model haqida
Ushbu model O'zbek tilidagi matnlarda Named Entity Recognition (NER) ni aniqlash uchun yaratilgan. Model turli xil kategoriyalardagi nomlangan entitetlarni aniqlashga qodir, jumladan shaxslar, joylar, tashkilotlar, sanalar va boshqalar. Ushbu model XLM-RoBERTa large arxitekturasiga asoslangan.

### Diqqat!!!
Model NEWS datasetda train qilingan va asosan NEWS textlardagi NER ni aniqlay olish aniqligi baland.

### Kategoriyalar
Model quyidagi NER kategoriyalarini aniqlashga qodir:
- **LOC (Joy nomlari)**
- **ORG (Tashkilot nomlari)**
- **PERSON (Shaxs nomlari)**
- **DATE (Sana ifodalari)**
- **MONEY (Pul miqdorlari)**
- **PERCENT (Foiz qiymatlari)**
- **QUANTITY (Miqdorlar)**
- **TIME (Vaqt ifodalari)**
- **PRODUCT (Mahsulot nomlari)**
- **EVENT (Voqea nomlari)**
- **WORK_OF_ART (San'at asarlari nomlari)**
- **LANGUAGE (Til nomlari)**
- **CARDINAL (Kardinal raqamlar)**
- **ORDINAL (Ordinall raqamlar)**
- **NORP (Millatlar yoki diniy/siyosiy guruhlar)**
- **FACILITY (Inshoot nomlari)**
- **LAW (Qonunlar yoki me'yorlar)**
- **GPE (Davlatlar, shaharlar, shtatlar)**

### Misollar
Model qanday ishlashini ko'rsatish uchun bir necha misollar:
```python
from transformers import pipeline, AutoTokenizer, AutoModelForTokenClassification

model_name_or_path = "sizning_model_yolingiz"
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForTokenClassification.from_pretrained(model_name_or_path).to("cuda")

nlp = pipeline("ner", model=model, tokenizer=tokenizer)

text = "Shavkat Mirziyoyev Rossiyada rasmiy safarda bo'ldi."
ner = nlp(text)

for entity in ner:
    print(entity)
```
Misol matni: "Shavkat Mirziyoyev Rossiyada rasmiy safarda bo'ldi."

Natijalar:
```python
[{'entity': 'B-PERSON', 'score': 0.88995147, 'index': 1, 'word': '▁Shavkat', 'start': 0, 'end': 7},
 {'entity': 'I-PERSON', 'score': 0.980681, 'index': 2, 'word': '▁Mirziyoyev', 'start': 8, 'end': 18},
 {'entity': 'B-GPE', 'score': 0.8208886, 'index': 3, 'word': '▁Rossiya', 'start': 19, 'end': 26}]
```

### Modelni yuklash va ishlatish
Modelni Hugging Face platformasidan yuklab olish va ishlatish uchun quyidagi koddan foydalanishingiz mumkin:
```python
from transformers import pipeline, AutoTokenizer, AutoModelForTokenClassification

model_name_or_path = "sizning_model_yolingiz"

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForTokenClassification.from_pretrained(model_name_or_path).to("cuda")

nlp = pipeline("ner", model=model, tokenizer=tokenizer)
```

## Bog'lanish

Agar savollaringiz bo'lsa yoki qo'shimcha ma'lumot kerak bo'lsa, iltimos biz bilan bog'laning.
LinkedIn: [Riskaliev Murad](https://www.linkedin.com/in/risqaliyevds/)

### Litsenziya
Ushbu model ochiq manba sifatida taqdim etiladi va barcha foydalanuvchilar uchun bepul foydalanish imkoniyatiga ega.

### Xulosa
O'zbek tili uchun NER modeli matnlarda turli xil nomlangan entitetlarni aniqlashda samarali yordam beradi. Modelning yuqori aniqligi va keng qamrovli kategoriyalari uni ilmiy tadqiqotlar, hujjatlarni tahlil qilish va boshqa ko'plab sohalarda qo'llash imkonini beradi.