 nagadomi
|
b7e116de54
Add support for multi GPU training (data parallel)
|
8 years ago |
|
nagadomi
|
d2cfb8f104
Add L1 criterion. Change the criterion of updating model
|
8 years ago |
|
nagadomi
|
3e77378983
Add SSIMCriterion (supports only forward())
|
8 years ago |
|
nagadomi
|
cabeeed2a7
Add Clip(0,1) to last layer
|
8 years ago |
|
nagadomi
|
382d493514
no confidence change for #128
|
9 years ago |
|
nagadomi
|
a0d8826779
Merge branch 'master' into upconv
|
9 years ago |
|
nagadomi
|
e50c7d1478
Remove unused modules
|
9 years ago |
|
nagadomi
|
af74a67bd1
Add -force_cudnn option; support for cuDNN in waifu2x.lua/web.lua
|
9 years ago |
|
nagadomi
|
8a65db7bab
Change the evaluation metric
|
9 years ago |
|
nagadomi
|
958dcfaeb8
Add more informative message for CUDA error
|
9 years ago |
|
nagadomi
|
5bfe504de1
Add error message when the CUDA module loading fails
|
9 years ago |
|
nagadomi
|
1900ac7500
Use PSNR for evaluation
|
9 years ago |
|
nagadomi
|
41581a0d55
Remove cleanup_model
|
9 years ago |
|
nagadomi
|
797b45ae23
Use roundf-like clip for 8 bit-depth image
|
9 years ago |
|
nagadomi
|
3b2aa67e55
Improve web interface
|
9 years ago |
|
nagadomi
|
9ddee6088d
Add WeightedHuberCriterion
|
9 years ago |
|
nagadomi
|
3abc5a03e3
refactor
|
9 years ago |
