 nagadomi
|
539941c234
Add support for url cache in web.lua
|
9 éve |
|
nagadomi
|
5db37f3392
Fix README syntax
|
9 éve |
|
nagadomi
|
7856ba90c7
Update trained models
|
9 éve |
|
nagadomi
|
268b9f6671
refactor tools/export_model.lua
|
9 éve |
|
nagadomi
|
6924920a2f
Fix tools/cudnn2cunn.lua no effect
|
9 éve |
|
nagadomi
|
903d945652
cleanup
|
9 éve |
|
nagadomi
|
2b9753157a
simplify
|
9 éve |
|
nagadomi
|
1f91548c6e
Fix model loading error when using ukbench model; Add error handling
|
9 éve |
|
nagadomi
|
1a20ec501b
Fix the evaluation metric in benchmark
|
9 éve |
|
nagadomi
|
44208f18ab
Fix handling for gamma embed PNG
|
9 éve |
|
nagadomi
|
bbd0ff39ad
Fix handling for CMYK JPEG
|
9 éve |
|
nagadomi
|
ee19e2608e
update README for dev
|
9 éve |
|
nagadomi
|
b35a9ae7d7
tuning
|
9 éve |
|
nagadomi
|
425898a3aa
Don't use cudnn.benchmark mode when predicting
|
9 éve |
|
nagadomi
|
490eb33a6b
Minimize the weighted huber loss instead of the weighted mean square error
|
9 éve |
|
nagadomi
|
243d8821be
Add -backend option to use cuDNN
|
9 éve |
|
nagadomi
|
9ddee6088d
Add WeightedHuberCriterion
|
9 éve |
|
nagadomi
|
9d63ed2947
Fix gradient scale in WeightedMSECriterion
|
9 éve |
|
nagadomi
|
4c691b4640
refactor
|
9 éve |
|
nagadomi
|
21ea5dd858
add cudnn converter
|
9 éve |
|
nagadomi
|
da786e15ba
remove noise_scale training
|
9 éve |
|
nagadomi
|
3abc5a03e3
refactor
|
9 éve |
|
nagadomi
|
cb81d21064
Add notice for v1.0 branch
|
9 éve |
|
nagadomi
|
956ae0d24c
merge from master
|
9 éve |
|
nagadomi
|
8dea362bed
sync from internal repo
|
9 éve |
|
nagadomi
|
36a53fa0c5
Update the torch7 install instruction in README
|
9 éve |
|
nagadomi
|
f399b34799
Added support for ru in nginx.conf
|
10 éve |
|
nagadomi
|
b35798016e
Added support for index.ru.html in web.lua
|
10 éve |
 vitaliylag
|
7aab0b4e12
typo
|
10 éve |
|
vitaliylag
|
dcba842b51
Update index.ru.html
|
10 éve |
