bcfefb492b
upd windows build magnet link
2020-12-11 18:26:13 +04:00
1ee7798fb9
.
2020-12-11 17:56:39 +04:00
704b5dc072
SAEHD: added new option Use AdaBelief optimizer?
...
Experimental AdaBelief optimizer. It requires more VRAM, but the accuracy of the model is higher, and lr_dropout is not needed.
2020-12-11 15:48:27 +04:00
e7d36b4287
leras: added AdaBelief optimizer
2020-12-11 15:47:47 +04:00
254a7cf5cf
fix xseg training
2020-12-11 15:47:11 +04:00
35945b257c
fix dimensions error
2020-12-11 14:34:37 +04:00
ca03bbca04
upd requirements-cuda.txt
2020-12-11 13:56:47 +04:00
764f3069a0
update magnet link
2020-12-11 11:47:27 +04:00
b9c9e7cffd
fix depth_to_space for tf2.4.0. Removing compute_output_shape in leras, because it uses CPU device, which does not support all ops.
2020-12-11 11:28:33 +04:00
bbf3a71a96
.
2020-12-02 00:55:27 +04:00
a03493c4f7
.
2020-12-02 00:54:47 +04:00
a35076a171
added new chinesse site
2020-11-28 09:53:31 +04:00
573ab6c22c
.
2020-11-25 12:20:15 +04:00
bb432b21f9
.
2020-11-25 12:19:43 +04:00
e9e7344424
upd
2020-11-22 18:26:54 +04:00
c516454566
fix rmsprop
2020-11-18 13:59:14 +04:00
874a7eba18
1
2020-11-18 13:41:19 +04:00
1adad3ece6
fix RMSprop.py
2020-11-18 13:38:27 +04:00
9ed662c522
DFLJPG, upd max save len
2020-11-16 11:04:03 +04:00
5446e17ea4
upd requirements
2020-11-13 23:12:14 +04:00
53a6df65af
1
2020-11-13 22:56:16 +04:00
0eb7e06ac1
upgrade to tf 2.4.0rc1
2020-11-13 17:00:07 +04:00
8b9b907682
upd readme
2020-11-11 22:12:20 +04:00
33f7ca5b62
upd readme
2020-11-11 22:11:43 +04:00
676310edc4
upd
2020-11-11 22:05:15 +04:00
b53a8e4372
upd readme
2020-11-04 16:12:19 +04:00
732c1e4ab2
readme.md: add Sassy Justice channel
2020-11-03 14:10:10 +04:00
6903841161
update face examples
2020-10-15 19:28:01 +04:00
b2f9ea8637
fix integer dimensions in model initialization
2020-10-03 22:38:29 +04:00
ce3f0676f0
readme: added RepresenUS youtube channel
2020-10-02 15:18:40 +04:00
d91e79fa64
README: added two new politician manipulation videos
2020-10-02 15:12:09 +04:00
4c47f0990d
fix magnet link
2020-09-15 09:11:43 +04:00
06e120a3df
upd magnet link
2020-09-15 09:08:07 +04:00
00b4f9090f
Fix mixed up links ( #885 )
2020-09-06 12:34:51 +04:00
768e56d862
fix readme
2020-09-02 20:00:12 +04:00
f4a661b742
fix readme
2020-08-05 23:05:44 +04:00
a61b7ee94d
fix README.md
2020-08-05 19:38:16 +04:00
b5bfec84c4
fix README.md
2020-08-05 19:36:56 +04:00
48168eb575
upd linux repo
2020-08-05 10:03:50 +04:00
3e7ee22ae3
Merger: fix load time of xseg if it has no model files
2020-07-29 00:41:53 +04:00
6134e57762
options_show_override{} for model summary in pretrain mode
2020-07-27 16:37:13 +04:00
1435dd3dd1
SAEHD: now random_warp is disabled for pretraining mode by default
2020-07-26 16:11:08 +04:00
ce95af9068
SAEHD: write_preview_history now works faster
...
The frequency at which the preview is saved now depends on the resolution.
For example 64x64 – every 10 iters. 448x448 – every 70 iters.
2020-07-18 11:03:13 +04:00
58670722dc
fix
2020-07-17 17:40:32 +04:00
b9b97861e1
fix mask for style losses
2020-07-17 17:34:07 +04:00
0efebe6bc5
upd deepfake_progress.png
2020-07-17 15:56:35 +04:00
3930073e32
Merger: added option “Number of workers?”
...
Specify the number of threads to process.
A low value may affect performance.
A high value may result in memory error.
The value may not be greater than CPU cores.
2020-07-17 15:46:01 +04:00
dd21880ecd
fix
2020-07-16 23:39:56 +04:00
6c7bb74ad4
README: update deepfake_progress, 448 pix
2020-07-16 22:26:43 +04:00
e8b04053e4
SAEHD:
...
Changed help for “Learning rate dropout” option:
When the face is trained enough, you can enable this option to get extra sharpness and reduce subpixel shake for less amount of iterations.
Enabled it before “disable random warp” and before GAN. n disabled. y enabled
cpu enabled on CPU. This allows not to use extra VRAM, sacrificing 20% time of iteration.
Changed help for GAN option:
Train the network in Generative Adversarial manner.
Forces the neural network to learn small details of the face.
Enable it only when the face is trained enough and don't disable.
Typical value is 0.1
improved GAN. Now it produces better skin detail, less patterned aggressive artifacts, works faster.
https://i.imgur.com/Nbh3mw1.png
2020-07-16 22:19:17 +04:00
iperov
