Update README.md
Browse files
README.md
CHANGED
|
@@ -3,7 +3,52 @@ language:
|
|
| 3 |
- tg
|
| 4 |
widget:
|
| 5 |
- text: "Пойтахти <mask> Душанбе"
|
| 6 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
---
|
| 8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
|
|
|
|
| 3 |
- tg
|
| 4 |
widget:
|
| 5 |
- text: "Пойтахти <mask> Душанбе"
|
| 6 |
+
tags:
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model_index:
|
| 9 |
+
- name: TajBERTo
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
name: Masked Language Modeling
|
| 13 |
+
type: fill-mask
|
| 14 |
---
|
| 15 |
|
| 16 |
+
# TajBERTo: RoBERTa-like Language model trained on Tajik
|
| 17 |
+
|
| 18 |
+
## Example pipeline
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
fill_mask = pipeline(
|
| 22 |
+
"fill-mask",
|
| 23 |
+
model="muhtasham/TajBERTo",
|
| 24 |
+
tokenizer="muhtasham/TajBERTo"
|
| 25 |
+
)
|
| 26 |
+
fill_mask("Пойтахти <mask> Душанбе")
|
| 27 |
+
|
| 28 |
+
# This is the beginning of a beautiful <mask>.
|
| 29 |
+
|
| 30 |
+
{'score': 0.1952248513698578,
|
| 31 |
+
'sequence': 'Пойтахти шаҳри Душанбе',
|
| 32 |
+
'token': 710,
|
| 33 |
+
'token_str': ' шаҳри'},
|
| 34 |
+
{'score': 0.029092855751514435,
|
| 35 |
+
'sequence': 'Пойтахти дар Душанбе',
|
| 36 |
+
'token': 310,
|
| 37 |
+
'token_str': ' дар'},
|
| 38 |
+
{'score': 0.020065447315573692,
|
| 39 |
+
'sequence': 'Пойтахти Душанбе Душанбе',
|
| 40 |
+
'token': 717,
|
| 41 |
+
'token_str': ' Душанбе'},
|
| 42 |
+
{'score': 0.016725927591323853,
|
| 43 |
+
'sequence': 'Пойтахти Тоҷикистон Душанбе',
|
| 44 |
+
'token': 424,
|
| 45 |
+
'token_str': ' Тоҷикистон'},
|
| 46 |
+
{'score': 0.011400512419641018,
|
| 47 |
+
'sequence': 'Пойтахти аз Душанбе',
|
| 48 |
+
'token': 335,
|
| 49 |
+
'token_str': ' аз'}
|
| 50 |
+
|
| 51 |
+
```
|
| 52 |
+
|
| 53 |
+
|
| 54 |
|