[Feature] Support Evaluation Hook of Translation Models #127
Conversation
Codecov Report
@@ Coverage Diff @@
## master #127 +/- ##
==========================================
- Coverage 75.28% 74.23% -1.06%
==========================================
Files 121 121
Lines 8189 8317 +128
Branches 1561 1601 +40
==========================================
+ Hits 6165 6174 +9
- Misses 1609 1726 +117
- Partials 415 417 +2
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
plyfager
changed the title
| @@ -136,10 +136,24 @@ | |||
| exp_name = 'cyclegan_summer2winter_id0' | |||
| work_dir = f'./work_dirs/experiments/{exp_name}' | |||
| # testA: 309, testB:238 | |||
| num_images = 238 | |||
There was a problem hiding this comment.
When would you change the file name to follow the convention of MMGEN?
There was a problem hiding this comment.
I will rename them and reorganize configs in #118 .
* support translation evaluation during training * fix lint * fix priority for eval hook * fix pr and add evaluation to configs * solve conflict Co-authored-by: yangyifei <PJLAB\yangyifei@shai14001042l.pjlab.org>
Support Evaluation Hook of Translation Models by
TranslationEvalHook.