github/DeepFaceLive
1
0
Fork 0
mirror of https://github.com/iperov/DeepFaceLive synced 2025-08-14 02:37:01 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

upd FAQ

Browse source
This commit is contained in:
iperov 2021-08-06 21:07:17 +04:00
commit 5ff19cb977
1 changed files with 16 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

19
doc/faq/faq.md
View file

@ -33,11 +33,24 @@ Src faceset is celebrity. Must be diverse enough.
Dst faceset is RTM WF faceset from the torrent.
AMP model settings:
224 res, WF, AE 256, inter 1024, e_dims 64, d_dims 64, d_mask_dims 22, morph factor 0.5, batch 8. Others by default.
Make a backup before every stage !
> Using SAEHD model.
res:224, WF, ae_dims:256, e_dims:64, d_dims:64, d_mask_dims 22, eyes_mouth_prio:Y, batch 8. Others are default.
1) enable pretrain mode. Train to 1kk
2) disable pretrain mode. Train to 1kk
3) lrd:N uniform_yaw:True, color_transfer:lct, train +500..800k
4) lrd:Y train +500..800k iters
5) disable random warp +500..800k iters
6) enable gan 0.1 gan_dims:32, train +100..300k iters
You can reuse this model to train new src faceset. In this case you should to delete inter_AB.npy from model folder, and train from stage 3.
> Using AMP model.
res:224, WF, ae_dims:256, inter_dims:1024, e_dims:64, d_dims:64, d_mask_dims:22, morph factor:0.5, batch 8. Others are default.
1) lrd:Y, train src-src for 1kk iters
2) delete inter_dst, lrd:N, color_transfer:lct, train +1kk
3) lrd:Y, train +1kk