CodeMirror/waifu2x
1
0
Fork 0
mirror of synced 2024-05-06 22:12:22 +12:00
Code Issues Releases Wiki Activity

rename srresnet_12l to resnet_14l because it's not the same as SRResNet

Browse source
This commit is contained in:
nagadomi 2017-01-21 16:53:10 +09:00
parent 0eccbc6555
commit 3447c6c44c
4 changed files with 6 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
appendix/benchmark.md
View file

@ -16,7 +16,7 @@ Urban100: https://github.com/jbhuang0604/SelfExSR
### 2x - PSNR
| Dataset/Model | Bicubic | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | srresnet_12l/photo |
| Dataset/Model | Bicubic | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | resnet_14l/photo |
|---------------|---------------|---------------|------------------|------------------|--------------------|
| BSD100 | 29.558 | 31.427 | 31.640 | 31.749 | 31.847 |
| Urban100 | 26.852 | 30.057 | 30.477 | 30.759 | 31.016 |
@ -25,21 +25,21 @@ Urban100: https://github.com/jbhuang0604/SelfExSR
Note: TTA is an ensemble technique that is supported by waifu2x. TTA method is 8x slower than non TTA method but it improves PSNR (~+0.1 on photo, ~+0.4 on art).
| Dataset/Model | Bicubic | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | srresnet_12l/photo |
| Dataset/Model | Bicubic | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | resnet_14l/photo |
|---------------|---------------|---------------|------------------|------------------|--------------------|
| BSD100 | 29.558 | 31.474 | 31.705 | 31.812 | 31.915 |
| Urban100 | 26.852 | 30.140 | 30.599 | 30.868 | 31.162 |
### 2x - benchmark elapsed time (sec)
| Dataset/Model | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | srresnet_12l/photo |
| Dataset/Model | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | resnet_14l/photo |
|---------------|---------------|------------------|------------------|--------------------|
| BSD100 | 4.057 | 2.509 | 4.947 | 6.86 |
| Urban100 | 16.349 | 7.083 | 14.178 | 27.87 |
### 2x with TTA - benchmark elapsed time (sec)
| Dataset/Model | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | srresnet_12l/photo |
| Dataset/Model | vgg\_7/photo | upconv\_7/photo | upconv\_7l/photo | resnet_14l/photo |
|---------------|---------------|------------------|------------------|--------------------|
| BSD100 | 36.611 | 20.219 | 42.486 | 60.38 |
| Urban100 | 132.416 | 65.125 | 129.916 | 255.20 |

4
lib/srcnn.lua
View file

@ -429,7 +429,7 @@ function srcnn.srresnet_2x(backend, ch)
end
-- large version of srresnet_2x. It's current best model but slow.
function srcnn.srresnet_12l(backend, ch)
function srcnn.resnet_14l(backend, ch)
local function resblock(backend, i, o)
local seq = nn.Sequential()
local con = nn.ConcatTable()
@ -463,7 +463,7 @@ function srcnn.srresnet_12l(backend, ch)
model:add(SpatialFullConvolution(backend, 256, ch, 4, 4, 2, 2, 3, 3):noBias())
model:add(w2nn.InplaceClip01())
model:add(nn.View(-1):setNumInputDims(3))
model.w2nn_arch_name = "srresnet_12l"
model.w2nn_arch_name = "resnet_14l"
model.w2nn_offset = 28
model.w2nn_scale_factor = 2
model.w2nn_resize = true

0
models/srresnet_12l/README.md → models/resnet_14l/README.md
View file

0
models/srresnet_12l/photo/scale2.0x_model.t7 → models/resnet_14l/photo/scale2.0x_model.t7
View file