Multilingual BART - (#3602)

- support mbart-en-ro weights
- add MBartTokenizer
This commit is contained in:
Sam Shleifer
2020-04-10 11:25:39 -04:00
committed by GitHub
parent f98d0ef2a2
commit 7a7fdf71f8
7 changed files with 232 additions and 38 deletions

View File

@@ -283,4 +283,7 @@ For a list that includes community-uploaded models, refer to `https://huggingfac
| +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
| | ``bart-large-cnn`` | | 12-layer, 1024-hidden, 16-heads, 406M parameters (same as base) |
| | | | bart-large base architecture finetuned on cnn summarization task |
| +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
| | ``mbart-large-en-ro`` | | 12-layer, 1024-hidden, 16-heads, 880M parameters |
| | | | bart-large architecture pretrained on cc25 multilingual data , finetuned on WMT english romanian translation. |
+-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+