norwegian-paws-x / translator.py
versae's picture
Create new file
1cf25f8
raw
history blame
1.65 kB
from functools import partial
import torch
from datasets import load_dataset
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
model_name = "facebook/nllb-200-3.3B" # "facebook/nllb-200-distilled-600M"
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = AutoModelForSeq2SeqLM.from_pretrained(model_name, use_auth_token=True, torch_dtype=torch.float32)
model.to(device, torch.float32, True)
tokenizer = AutoTokenizer.from_pretrained(
model_name, use_auth_token=True, src_lang="eng_Latn"
)
def to_lang_code(text, lang_code):
inputs = tokenizer(text, return_tensors="pt").to(device)
translated_tokens = model.generate(
**inputs,
forced_bos_token_id=tokenizer.lang_code_to_id[lang_code],
max_length=int(len(inputs.tokens()) * 1.5) # 50% more tokens for the translation just in case
)
return tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
if __name__ == "__main__":
languages = (("nb", "nob_Latn"), ("nn", "nno_Latn"))
ds = load_dataset("paws-x", "en")
dss = {}
for lang, translate_code in languages:
translate = partial(to_lang_code, lang_code=translate_code)
dss[lang] = ds.map(lambda example: {
"sentence1": translate(example["sentence1"]),
"sentence2": translate(example["sentence2"]),
}, desc=f"Translating to {lang}")
for split in ("test", "validation", "train"):
json_lines = dss[lang][split].to_pandas().to_json(orient='records', lines=True)
with open(f"{lang}_{split}.json", "w") as json_file:
json_file.write(json_lines)