Views
No views yet
1from transformers import NllbTokenizer, AutoModelForSeq2SeqLM
2
3tokenizer = NllbTokenizer.from_pretrained("CenIA/nllb-200-3.3B-spa-rap")
4model = AutoModelForSeq2SeqLM.from_pretrained("CenIA/nllb-200-3.3B-spa-rap")
5
6def translate(sentence: str, translate_from="spa_Latn", translate_to="mri_Latn") -> str:
7 tokenizer.src_lang = translate_from
8 tokenizer.tgt_lang = translate_to
9
10 inputs = tokenizer(sentence, return_tensors="pt")
11 result = model.generate(**inputs, forced_bos_token_id=tokenizer.convert_tokens_to_ids(translate_to))
12 decoded = tokenizer.batch_decode(result, skip_special_tokens=True)[0]
13
14 return decoded
15
16traduction = translate("Hola, ¿cómo estás?")
17
18print(traduction)