update training instruction

This commit is contained in:
iperov 2022-08-18 00:06:52 +04:00
commit 09ef6d3096

View file

@ -59,11 +59,11 @@ Gather 5000+ samples of your face with various conditions using webcam which wil
Here public storage https://disk.yandex.ru/d/7i5XTKIKVg5UUg with facesets and models. Here public storage https://disk.yandex.ru/d/7i5XTKIKVg5UUg with facesets and models.
> Using pretrained "RTT model 224.zip" from public storage (see above) > Using pretrained "RTT model 224 V2.zip" from public storage (see above)
Make a backup before every stage ! Make a backup before every stage !
1. place RTM WF Faceset from public storage (see above) to workspace/data_dst/aligned 1. place RTM WF Faceset V2 from public storage (see above) to workspace/data_dst/aligned
2. place your celeb to workspace/data_src/aligned 2. place your celeb to workspace/data_src/aligned
@ -71,21 +71,21 @@ Make a backup before every stage !
4. replace dst faceset with your faceset in workspace/data_dst/aligned 4. replace dst faceset with your faceset in workspace/data_dst/aligned
5. continue train +500.000 5. continue train +500.000, (optional) deleting inter_AB.npy every 100.000 (save, delete, continue run)
6. random_warp:OFF, train +500.000 6. random_warp:OFF, train +500.000
7. GAN 0.1 power, patch size 28, gan_dims:32. Train until the src loss value has not increased in the last 12 hours. 7. GAN 0.1 power, patch size 28, gan_dims:32. Train until the src loss value has not increased in the last 12 hours.
8. finalize model by disabling masked training for 100-200 (not thousand) iterations. 8. (optional) finalize model by disabling masked training for 100-200 (not thousand) iterations.
> Using SAEHD model from scratch. > Using SAEHD model from scratch.
res:224, WF, archi:liae-udt, ae_dims:512, e_dims:64, d_dims:64, d_mask_dims:32, eyes_mouth_prio:Y, blur_out_mask:Y, uniform_yaw:Y, lr_dropout:Y, batch:8. Others by default. res:224, WF, archi:liae-udt, ae_dims:512, e_dims:64, d_dims:64, d_mask_dims:32, eyes_mouth_prio:N, blur_out_mask:Y, uniform_yaw:Y, lr_dropout:Y, batch:8. Others by default.
Make a backup before every stage ! Make a backup before every stage !
1. place RTM WF Faceset from public storage (see above) to workspace/data_dst/aligned 1. place RTM WF Faceset V2 from public storage (see above) to workspace/data_dst/aligned
2. place your celeb to workspace/data_src/aligned 2. place your celeb to workspace/data_src/aligned
@ -101,7 +101,7 @@ Make a backup before every stage !
8. GAN 0.1 power, gan_dims:32, Train until the src loss value has not increased in the last 12 hours. 8. GAN 0.1 power, gan_dims:32, Train until the src loss value has not increased in the last 12 hours.
9. finalize model by disabling masked training for 100-200 (not thousand) iterations. 9. (optional) finalize model by disabling masked training for 100-200 (not thousand) iterations.
10. export the model in .dfm format for use in DeepFaceLive. You can also try ordering a deepfake model from someone in Discord or forum. 10. export the model in .dfm format for use in DeepFaceLive. You can also try ordering a deepfake model from someone in Discord or forum.
@ -119,31 +119,31 @@ Src faceset is celebrity. Must be diverse enough in yaw, light and shadow condit
Do not mix different age. The best result is obtained when the face is filmed from a short period of time and does not change the makeup and structure. Do not mix different age. The best result is obtained when the face is filmed from a short period of time and does not change the makeup and structure.
Src faceset should be xseg'ed and applied. You can apply Generic XSeg to src faceset. Src faceset should be xseg'ed and applied. You can apply Generic XSeg to src faceset.
> Using pretrained "RTT model 224.zip" from public storage (see above) > Using pretrained "RTT model 224 V2.zip" from public storage (see above)
Make a backup before every stage ! Make a backup before every stage !
1. place RTM WF Faceset from public storage (see above) to workspace/data_dst/aligned 1. place RTM WF Faceset V2 from public storage (see above) to workspace/data_dst/aligned
2. place your celeb to workspace/data_src/aligned 2. place your celeb to workspace/data_src/aligned
3. place model folder to workspace/model 3. place model folder to workspace/model
4. do not change settings, train +500.000 iters 4. do not change settings, train +500.000 iters, + (optional) deleting inter_AB.npy every 100.000 (save, delete, continue run)
5. random_warp OFF, train +500.000, periodically (every 100.000 iters) disable masked training for 5.000 iters and enable again 5. random_warp OFF, train +500.000, + (optional) periodically (every 100.000 iters) disable masked training for 5.000 iters and enable again
6. GAN 0.1 power, patch size 28, gan_dims:32. Train until the src loss value has not increased in the last 12 hours. 6. GAN 0.1 power, patch size 28, gan_dims:32. Train until the src loss value has not increased in the last 12 hours.
7. finalize model by disabling masked training for 100-200 (not thousand) iterations. 7. (optional) finalize model by disabling masked training for 100-200 (not thousand) iterations.
> Using SAEHD model from scratch > Using SAEHD model from scratch
res:224, WF, archi:liae-udt, ae_dims:512, e_dims:64, d_dims:64, d_mask_dims:32, eyes_mouth_prio:Y, blur_out_mask:Y, uniform_yaw:Y, lr_dropout:Y, batch:8. Others by default. res:224, WF, archi:liae-udt, ae_dims:512, e_dims:64, d_dims:64, d_mask_dims:32, eyes_mouth_prio:N, blur_out_mask:Y, uniform_yaw:Y, lr_dropout:Y, batch:8. Others by default.
Make a backup before every stage ! Make a backup before every stage !
1. place RTM WF Faceset from public storage (see above) to workspace/data_dst/aligned 1. place RTM WF Faceset V2 from public storage (see above) to workspace/data_dst/aligned
2. place your celeb to workspace/data_src/aligned 2. place your celeb to workspace/data_src/aligned
@ -155,7 +155,7 @@ Make a backup before every stage !
6. GAN 0.1 power, gan_dims:32. Train until the src loss value has not increased in the last 12 hours. 6. GAN 0.1 power, gan_dims:32. Train until the src loss value has not increased in the last 12 hours.
7. finalize model by disabling masked training for 100-200 (not thousand) iterations. 7. (optional) finalize model by disabling masked training for 100-200 (not thousand) iterations.
> reusing trained SAEHD RTM model > reusing trained SAEHD RTM model