default param values for sacrebleu

[base-y]

I see default params for sacrebleu as smooth_method=exp and tokenize=None (uses 13a default tokenizer) while in sacrebleu python lib (https://github.com/mjpost/sacrebleu), the default params seem to be smooth_method=floor and tokenize='none' (which uses BaseTokenizer).

Which set is officially recognized and why was exp smooth method chosen in hf datasets?

[mariosasko]

Hi! The default parameters should match since our implementation calls sacrebleu.corpus_blue, which has the following signature: https://github.com/mjpost/sacrebleu/blob/8e7abf591f6acff69dcbcad8d7af94754296fcb4/sacrebleu/compat.py#L9-L16 (notice smooth-methd='exp' and tokenize=BLEU.TOKENIZER_DEFAULT). And if you follow the code further, you can see that tokenize=None is equal to tokenize=BLEU.TOKENIZER_DEFAULT.