Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
- ta
|
5 |
+
license: cc-by-4.0
|
6 |
+
tags:
|
7 |
+
- translation
|
8 |
+
- tamil
|
9 |
+
- colloquial-tamil
|
10 |
+
- fine-tuned
|
11 |
+
- text-to-text
|
12 |
+
|
13 |
+
|
14 |
+
|
15 |
+
## 🔥 Example Usage
|
16 |
+
|
17 |
+
Load and test the model using **Hugging Face Transformers**:
|
18 |
+
|
19 |
+
```python
|
20 |
+
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
21 |
+
# Load model and tokenizer
|
22 |
+
model_name = "aishu15/colloquial-tamil"
|
23 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
24 |
+
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
|
25 |
+
# Function to translate text
|
26 |
+
def translate(text):
|
27 |
+
inputs = tokenizer(text, return_tensors="pt")
|
28 |
+
outputs = model.generate(**inputs, max_length=128)
|
29 |
+
return tokenizer.decode(outputs[0], skip_special_tokens=True)
|
30 |
+
# Example translations
|
31 |
+
test_sentences = [
|
32 |
+
"This is so beautiful",
|
33 |
+
"Bro, are you coming or not?",
|
34 |
+
"My mom is gonna kill me if I don't reach home now!"
|
35 |
+
]
|
36 |
+
for sentence in test_sentences:
|
37 |
+
print(f"English: {sentence}")
|
38 |
+
print(f"Colloquial Tamil: {translate(sentence)}\n")
|