Update checkpoint for transformers>=4.29

#2
by ArthurZ HF staff - opened
Language Technology Research Group at the University of Helsinki org

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-base-zle-bat')
>>> inputs = tokenizer("' >>en<< Hey how are you?'", return_tensors="pt", padding=True)
>>> model = MarianMTModel.from_pretrained('Helsinki-NLP/opus-mt-tc-base-zle-bat')
>>> print(tokenizer.batch_decode(model.generate(**inputs)))
['<pad> '' how are you?' "</s>']
Language Technology Research Group at the University of Helsinki org

Automatically merging the PR.

ArthurZ changed pull request status to merged

Sign up or log in to comment