Views
No views yet
1from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
2
3# Load model
4model_name = "YOUR_USERNAME/multilingual-news-translator"
5model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
6tokenizer = AutoTokenizer.from_pretrained(model_name)
7
8# Translate to Hindi
9text = "Global markets showed strong growth today"
10tokenizer.src_lang = "eng_Latn"
11inputs = tokenizer(text, return_tensors="pt")
12outputs = model.generate(
13 **inputs,
14 forced_bos_token_id=tokenizer.lang_code_to_id["hin_Deva"],
15 max_length=512
16)
17translation = tokenizer.decode(outputs[0], skip_special_tokens=True)
18print(translation)| Language | Code | Script |
|---|---|---|
| English | eng_Latn | Source language |
| Hindi | hin_Deva | देवनागरी |
| Telugu | tel_Telu | తెలుగు |
| Tamil | tam_Taml | தமிழ் |
| Kannada | kan_Knda | ಕನ್ನಡ |
| Bengali | ben_Beng | বাংলা |
| Malayalam | mal_Mlym | മലയാളം |
| Spanish | spa_Latn | Latin |
| French | fra_Latn | Latin |
| Japanese | jpn_Jpan | 日本語 |
| Chinese | zho_Hans | 简体中文 |
1from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
2
3class NewsTranslator:
4 def __init__(self, model_name):
5 self.tokenizer = AutoTokenizer.from_pretrained(model_name)
6 self.model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
7 self.languages = {
8 'hindi': 'hin_Deva',
9 'tamil': 'tam_Taml',
10 'spanish': 'spa_Latn',
11 'french': 'fra_Latn'
12 }
13
14 def translate(self, text, target_lang):
15 self.tokenizer.src_lang = "eng_Latn"
16 inputs = self.tokenizer(text, return_tensors="pt", truncation=True)
17 outputs = self.model.generate(
18 **inputs,
19 forced_bos_token_id=self.tokenizer.lang_code_to_id[self.languages[target_lang]],
20 max_length=512,
21 num_beams=5
22 )
23 return self.tokenizer.decode(outputs[0], skip_special_tokens=True)
24
25# Usage
26translator = NewsTranslator("YOUR_USERNAME/multilingual-news-translator")
27result = translator.translate("Breaking news from around the world", "hindi")
28print(result)