 nagadomi
|
25e293202a
Add support for tta_level=1; Add support for TTA to web.lua
|
9 年之前 |
|
nagadomi
|
af74a67bd1
Add -force_cudnn option; support for cuDNN in waifu2x.lua/web.lua
|
9 年之前 |
|
nagadomi
|
599da6a665
refactor
|
9 年之前 |
|
nagadomi
|
6be1479710
Fix a block_size issue when using upconv_7 model
|
9 年之前 |
|
nagadomi
|
9103d393fe
Fix a performance issue
|
9 年之前 |
|
nagadomi
|
83188c5ab7
Add support for method=noise_scale to tools/benchmark.lua
|
9 年之前 |
|
nagadomi
|
c16d0a07a2
Add -crop_size and -batch_size option to tools/benchmark.lua. Fix a bug in tta mode.
|
9 年之前 |
|
nagadomi
|
b8ff8c6787
Remove -gamma_correction option
|
9 年之前 |
|
nagadomi
|
01b2e6d441
Remove -upsampling_filter option
|
9 年之前 |
|
nagadomi
|
afac4b52ab
Add -batch_size option to waifu2x.lua/web.lua
|
9 年之前 |
|
nagadomi
|
0b949c05a7
Add support for TTA level
|
9 年之前 |
|
nagadomi
|
9514027f65
Fix a bug that -nr_rate is not used
|
9 年之前 |
|
nagadomi
|
e5cfd3dfce
Add -resume option
|
9 年之前 |
|
nagadomi
|
37bc7a5eea
Add support for new noise_scale method
|
9 年之前 |
|
nagadomi
|
6c758ec5c0
Correct messages
|
9 年之前 |
|
nagadomi
|
51914b894a
change weight initialization and upconv_7
|
9 年之前 |
|
nagadomi
|
307ae40883
Add noise_scale training
|
9 年之前 |
|
nagadomi
|
3b09bff8cf
randomly swap fg/bg color in dots/gen
|
9 年之前 |
|
nagadomi
|
d0630d3a20
individual filters and box-only support
|
9 年之前 |
|
nagadomi
|
5e222a3981
Add -tta and -resize_blur option to benchmark
|
9 年之前 |
|
nagadomi
|
abae4cb855
Fix processing time
|
9 年之前 |
|
nagadomi
|
0349fc774c
refactor
|
9 年之前 |
|
nagadomi
|
c89fd7249a
Add learning_rate_decay
|
9 年之前 |
|
nagadomi
|
70eb2b508f
Fix a performance problem in resampling
|
9 年之前 |
|
nagadomi
|
70a2849e39
Fix missing file
|
9 年之前 |
|
nagadomi
|
634046d5f0
Fix training mode
|
9 年之前 |
|
nagadomi
|
b96bc5d453
Use correct criterion
|
9 年之前 |
|
nagadomi
|
99e6dd1a57
Fix border removing
|
9 年之前 |
|
nagadomi
|
8a65db7bab
Change the evaluation metric
|
9 年之前 |
|
nagadomi
|
8088460a20
Add oracle_rate option
|
9 年之前 |
