Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,35 @@
|
|
1 |
-
---
|
2 |
-
license: cc-by-nc-4.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
datasets:
|
4 |
+
- guymorlan/levanti
|
5 |
+
language:
|
6 |
+
- ar
|
7 |
+
- he
|
8 |
+
pipeline_tag: translation
|
9 |
+
widget:
|
10 |
+
- text: P אני רוצה ללכת מחר לחנות
|
11 |
+
---
|
12 |
+
|
13 |
+
# Levanti Hebrew -> colloquial Levantine Arabic translator
|
14 |
+
|
15 |
+
Trained on the [Levanti](https://huggingface.co/datasets/guymorlan/levanti) dataset by fine-tuning [Helsinki-NLP/opus-mt-he-ar](https://huggingface.co/Helsinki-NLP/opus-mt-ar-he) for 8 epochs.
|
16 |
+
This model is trained to support dialect conditional generation by utilizing the first token (followed by a space) as an indicator of the desired dialect:
|
17 |
+
* **P** for Palestinian
|
18 |
+
* **L** for Lebanese
|
19 |
+
* **S** for Syrian
|
20 |
+
* **E** for Egyptian
|
21 |
+
|
22 |
+
# Example usage
|
23 |
+
|
24 |
+
```python
|
25 |
+
from transformers import pipeline
|
26 |
+
trans = pipeline("translation", "guymorlan/levanti_translate_he_ar")
|
27 |
+
trans("P אני רוצה ללכת מחר לחנות")
|
28 |
+
```
|
29 |
+
```
|
30 |
+
Out[1]: [{'translation_text': 'بدي أروح ع الدكان بكرا'}]
|
31 |
+
```
|
32 |
+
|
33 |
+
# Attribution
|
34 |
+
Created by Guy Mor-Lan.<br>
|
35 |
+
Contact: guy.mor AT mail.huji.ac.il
|