Timeline for Should the data be shuffled on a translation dataset
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 17, 2019 at 8:17 | comment | added | noe | You can check the marian examples page for plenty of examples. Search for post-editing in that page and you will find the dual encoder thing. The gradient checkpointing is a new addition and I don't know of any documentation yet. | |
| Dec 16, 2019 at 12:11 | comment | added | user86834 | Nice! Can you may send me a link, or some command line options? | |
| Dec 16, 2019 at 0:04 | comment | added | noe | Marian supports long sentences (i.e. whole documents) via gradient checkpointing and auxiliary encoders (normally used for automatic post-editing). | |
| Dec 15, 2019 at 17:11 | comment | added | user86834 | "In document-level MT you normally have context sentences that are either ingested with an auxiliary encoder" <- Can you maybe pinpoint me to an implementation concerning that or to a paper? | |
| Dec 15, 2019 at 17:07 | vote | accept | user86834 | ||
| Dec 13, 2019 at 16:55 | history | answered | noe | CC BY-SA 4.0 |