You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Oct 31, 2023. It is now read-only.
The paper shows the best en-fr bleu is 33.4. The readme.md shows
'epoch -> 7
valid_fr-en_mt_bleu -> 28.36
valid_en-fr_mt_bleu -> 30.50
test_fr-en_mt_bleu -> 34.02
test_en-fr_mt_bleu -> 36.62'.
Does this result from the max_len parameter which removes the long sentences from parallel test corpus?
The text was updated successfully, but these errors were encountered:
No, the difference comes from the fact that the monolingual dataset is different. In the paper we use all NewsCrawl, in the Github we just use NewsCrawl 2013 and 2014 I believe, which is more in domain with newstest2014 on which we evaluate.
The paper shows the best en-fr bleu is 33.4. The readme.md shows
'epoch -> 7
valid_fr-en_mt_bleu -> 28.36
valid_en-fr_mt_bleu -> 30.50
test_fr-en_mt_bleu -> 34.02
test_en-fr_mt_bleu -> 36.62'.
Does this result from the max_len parameter which removes the long sentences from parallel test corpus?
The text was updated successfully, but these errors were encountered: