Update README.md
Browse files
README.md
CHANGED
@@ -9,4 +9,22 @@ base_model:
|
|
9 |
---
|
10 |
|
11 |
mt-5-base model fine tuned for restoration and recapitalization for Macedonian language.
|
12 |
-
The model is fine-tuned on a subset of the Macedonian portion of Wikipedia.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
mt-5-base model fine tuned for restoration and recapitalization for Macedonian language.
|
12 |
+
The model is fine-tuned on a subset of the Macedonian portion of Wikipedia.
|
13 |
+
|
14 |
+
|
15 |
+
## Usage
|
16 |
+
|
17 |
+
```python
|
18 |
+
from transformers import T5Tokenizer, T5ForConditionalGeneration
|
19 |
+
recap_model_name = "Macedonian-ASR/mt5-restore-capitalization-macedonian"
|
20 |
+
recap_tokenizer = T5Tokenizer.from_pretrained(recap_model_name)
|
21 |
+
recap_model = T5ForConditionalGeneration.from_pretrained(recap_model_name)
|
22 |
+
recap_model.to(device)
|
23 |
+
|
24 |
+
sentence = "скопје е главен град на македонија"
|
25 |
+
inputs = recap_tokenizer(["restore capitalization and punctuation: " + sentence], return_tensors="pt", padding=True).to(device)
|
26 |
+
outputs = recap_model.generate(**inputs, max_length=768, num_beams=5, early_stopping=True).squeeze(0)
|
27 |
+
recap_result = recap_tokenizer.decode(outputs, skip_special_tokens=True)
|
28 |
+
print(recap_result)
|
29 |
+
-> Скопје е главен град на Македонија.
|
30 |
+
```
|