bug when using command-line tool "fairseq-score" with argument "--sentence-bleu"
To Reproduce
Steps to reproduce the behavior (always include the command you ran):
Run cmd > fairseq-score -s sys -r ref --sentence-bleu
See error
Error messages
File "fairseq\fairseq_cli\score.py", line 67, in score
scorer = bleu.Scorer(dict.pad(), dict.eos(), dict.unk())
TypeError: Scorer.init() takes 2 positional arguments but 4 were given
Additional context
the "bleu.Scorer.init()" takes "BleuConfig", but "dict.pad(), dict.eos(), dict.unk()" are given in farirseq_cli/score.py
🐛 Bug
bug when using command-line tool "fairseq-score" with argument "--sentence-bleu"
To Reproduce
Steps to reproduce the behavior (always include the command you ran):
Error messages
File "fairseq\fairseq_cli\score.py", line 67, in score scorer = bleu.Scorer(dict.pad(), dict.eos(), dict.unk()) TypeError: Scorer.init() takes 2 positional arguments but 4 were given
Additional context
the "bleu.Scorer.init()" takes "BleuConfig", but "dict.pad(), dict.eos(), dict.unk()" are given in farirseq_cli/score.py