Helsinki-NLP

/

opus-mt-tc-big-gmq-zlw

like 0

Follow

Language Technology Research Group at the University of Helsinki 233

Following the merge of a PR in transformers it appeared that this model was not properly converted. This PR will fix the inference and was tested using the following script:

>>> from transformers import AutoTokenizer, MarianMTModel
>>> tokenizer = AutoTokenizer.from_pretrained('Helsinki-NLP/opus-mt-tc-big-gmq-zlw')
>>> inputs = tokenizer("' >>en<< Hey how are you?'", return_tensors="pt", padding=True)
>>> model = MarianMTModel.from_pretrained('Helsinki-NLP/opus-mt-tc-big-gmq-zlw')
>>> print(tokenizer.batch_decode(model.generate(**inputs)))
['<pad> "Hho how how are you?" "Hho how a are you? "? "? "ho ho ho a a a a you you??? "? "? "???? " "ho ho a a you? "????????? "? "?'ho ho ho " " a " " " " " " "? " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "</s>']