Upload README (1).md
Browse files- README (1).md +34 -0
README (1).md
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
Bu model https://github.com/stefan-it/turkish-bert'in; aşağıdaki 5 kategorinin olduğu metin sınıflandırma verilerine göre fine-tuned edilmiş halidir.
|
| 3 |
+
|
| 4 |
+
code_to_label={
|
| 5 |
+
|
| 6 |
+
'LABEL_0': 'INSULT ',
|
| 7 |
+
|
| 8 |
+
'LABEL_1': 'RACIST ',
|
| 9 |
+
|
| 10 |
+
'LABEL_2': 'SEXIST',
|
| 11 |
+
|
| 12 |
+
'LABEL_3': 'PROFANITY ',
|
| 13 |
+
|
| 14 |
+
'LABEL_4': 'OTHER' }
|
| 15 |
+
|
| 16 |
+
```python
|
| 17 |
+
|
| 18 |
+
from transformers import pipeline, AutoModelForTokenClassification, AutoTokenizer, AutoModelForSequenceClassification
|
| 19 |
+
tokenizer= AutoTokenizer.from_pretrained("ennp/bert-turkish-text-classification-cased")
|
| 20 |
+
|
| 21 |
+
model= AutoModelForSequenceClassification.from_pretrained("ennp/bert-turkish-text-classification-cased")
|
| 22 |
+
|
| 23 |
+
nlp=pipeline("sentiment-analysis", model=model, tokenizer=tokenizer)
|
| 24 |
+
|
| 25 |
+
code_to_label={
|
| 26 |
+
'LABEL_0': 'INSULT ',
|
| 27 |
+
'LABEL_1': 'RACIST ',
|
| 28 |
+
'LABEL_2': 'SEXIST',
|
| 29 |
+
'LABEL_3': 'PROFANITY ',
|
| 30 |
+
'LABEL_4': 'OTHER' }
|
| 31 |
+
|
| 32 |
+
code_to_label[nlp("kıl herif gibi davranma")[0]['label']]
|
| 33 |
+
```
|
| 34 |
+
|