Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,31 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
tags:
|
| 3 |
+
|
| 4 |
+
- text2text-generation
|
| 5 |
+
|
| 6 |
+
- Trinidadian Creole
|
| 7 |
+
|
| 8 |
+
- Caribbean dialect
|
| 9 |
+
|
| 10 |
license: apache-2.0
|
| 11 |
+
|
| 12 |
---
|
| 13 |
+
|
| 14 |
+
# Standard English to Trinidad English Creole Translator
|
| 15 |
+
This model utilises T5-base pre-trained model. It was fine tuned using a custom dataset for translation of English to Trinidad English Creole. This model will be updated periodically as more data is compiled. For more on the Caribbean English Creole checkout the library [Caribe](https://pypi.org/project/Caribe/).
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
___
|
| 20 |
+
# Usage with Transformers
|
| 21 |
+
```python
|
| 22 |
+
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
| 23 |
+
tokenizer = AutoTokenizer.from_pretrained("KES/ENG-TEC")
|
| 24 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("KES/ENG-TEC")
|
| 25 |
+
text = "Where you going now?"
|
| 26 |
+
inputs = tokenizer("eng:"+text, truncation=True, return_tensors='pt')
|
| 27 |
+
output = model.generate(inputs['input_ids'], num_beams=4, max_length=512, early_stopping=True)
|
| 28 |
+
translation=tokenizer.batch_decode(output, skip_special_tokens=True)
|
| 29 |
+
print("".join(translation)) #translation: Weh yuh going now.
|
| 30 |
+
```
|
| 31 |
+
___
|