 nagadomi
|
66bff68ef1
split first
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
06e08253e4
Fix some multhread bug
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
a14e6acec3
mutex
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
5a3d012f4e
Add random blur method for data augmentation
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
451ee1407f
Stop calculate the instance loss when oracle_rate=0
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
33e6bc888e
Fix the number of threads
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
c2e4bb4380
Support for multi-thread in training
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
7ffd8a51fd
fix message
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
3a70d09921
create new directory when the specified directory is not found
|
%!s(int64=8) %!d(string=hai) anos |
|
nagadomi
|
81df729a8a
Remove -loss option
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
edac608f18
Add support for user specified pairwise data for universal filter
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
de669d24a4
Change default -validation_crops to 200.
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
b8ff8c6787
Remove -gamma_correction option
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
01b2e6d441
Remove -upsampling_filter option
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
e5cfd3dfce
Add -resume option
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
6c758ec5c0
Correct messages
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
307ae40883
Add noise_scale training
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
d0630d3a20
individual filters and box-only support
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
c89fd7249a
Add learning_rate_decay
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
634046d5f0
Fix training mode
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
b96bc5d453
Use correct criterion
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
8088460a20
Add oracle_rate option
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
8fec6f1b5a
Change the learning rate decay rate
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
7814691cbf
Add resize_blur parameter
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
a210090033
Convert model files; Add new pretrained model
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
51ae485cd1
Add new models
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
b8088ca209
Remove the limit of learning_rate_decay
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
7af5c9443d
Add model option and 12 layers net
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
68a6d4cef5
Use MSE instead of PSNR
|
%!s(int64=9) %!d(string=hai) anos |
|
nagadomi
|
fa9355be7c
Fix validation metric
|
%!s(int64=9) %!d(string=hai) anos |
