Update README.md
Browse files
README.md
CHANGED
|
@@ -11,6 +11,10 @@ tags:
|
|
| 11 |
- chats
|
| 12 |
- embeddings
|
| 13 |
- coherence
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
---
|
| 15 |
# Model Card
|
| 16 |
|
|
@@ -26,7 +30,7 @@ The task perfectly aligns with the Next Sentence Prediction task, so the fine-tu
|
|
| 26 |
**Note:** if two messages have `reply_to` relationship, then **they have "zero" label**. This is because of the NSP formulation.
|
| 27 |
```python
|
| 28 |
from transformers import AutoTokenizer, BertForNextSentencePrediction
|
| 29 |
-
tokenizer = AutoTokenizer.from_pretrained("rubert_reply_recovery", )
|
| 30 |
model = BertForNextSentencePrediction.from_pretrained("rubert_reply_recovery", )
|
| 31 |
|
| 32 |
inputs = tokenizer(['Где можно получить СНИЛС?', 'Я тут уже много лет'], ["Можете в МФЦ", "Куда отправить это письмо?"], return_tensors='pt',
|
|
|
|
| 11 |
- chats
|
| 12 |
- embeddings
|
| 13 |
- coherence
|
| 14 |
+
widget:
|
| 15 |
+
- text: Где можно получить СНИЛС? Можете в МФЦ
|
| 16 |
+
- text: Я тут уже много лет Куда отправить это письмо?
|
| 17 |
+
pipeline_tag: text-classification
|
| 18 |
---
|
| 19 |
# Model Card
|
| 20 |
|
|
|
|
| 30 |
**Note:** if two messages have `reply_to` relationship, then **they have "zero" label**. This is because of the NSP formulation.
|
| 31 |
```python
|
| 32 |
from transformers import AutoTokenizer, BertForNextSentencePrediction
|
| 33 |
+
tokenizer = AutoTokenizer.from_pretrained("astromis/rubert_reply_recovery", )
|
| 34 |
model = BertForNextSentencePrediction.from_pretrained("rubert_reply_recovery", )
|
| 35 |
|
| 36 |
inputs = tokenizer(['Где можно получить СНИЛС?', 'Я тут уже много лет'], ["Можете в МФЦ", "Куда отправить это письмо?"], return_tensors='pt',
|