Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Add model config used in pr6 benchmarks
- Loading branch information
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -76,11 +76,22 @@ | |
| "/vocab.esen.spm", | ||
| "/vocab.esen.spm" | ||
| ], | ||
| "beam-size": 1, | ||
| "mini-batch": 32, | ||
| "maxi-batch": 100, | ||
| "maxi-batch-sort": "src", | ||
This comment has been minimized.
Sorry, something went wrong.
This comment has been minimized.
Sorry, something went wrong. 
motin
Author
Contributor
|
||
| "workspace": 128, | ||
| "skip-cost": true, | ||
| "cpu-threads": 1, | ||
| "shortlist": [ | ||
| `/lex.${lang}.s2t`, | ||
| 50, | ||
| 50, | ||
| ] | ||
| // TODO: Enable when wormhole is enabled | ||
| // "int8shift": true, | ||
| // TODO: Enable when loading of binary models is supported and we use model.intgemm.alphas.bin | ||
| // "int8shiftAlphaAll": true, | ||
| }; | ||
|
|
||
| // Instantiate the TranslationModel | ||
|
|
||
If you're using the API currently in place,
mini-batch,maxi-batch,maxi-batch-sortare unused. If there are benchmarks being run and reported, they may not exactly be comparable.--max-input-sentence-tokensandmax-input-tokensare present at the moment.max-input-tokenswill probably get renamed tomini-batch-words, to align with it's counter-part inmarian-decoder.I'm sorry for this mess, I expect to clean this up eventually. Just letting you know a possible issue.