DeepFaceLab is the leading software for creating deepfakes.
GNU General Public License v3.0
16.38k
stars
30
forks
source link
The Model seem to have a problem to try to fit with the face on the Deepfacelive , but I tried other models it does look good , might be the problem on the training state , can some one help me how to avoid it ? #5691
I'm trying to train my own models to learn how it work , and encounter some problem with the result and I can't figure it out where the problem is , is it the Xseg mask ? and here some of the setting I'm running on ================== Model Summary =================== == == == Model name: new_SAEHD == == == == Current iteration: 3207409 == == == ==---------------- Model Options -----------------== == == == resolution: 224 == == face_type: wf == == models_opt_on_gpu: True == == archi: liae-udt == == ae_dims: 512 == == e_dims: 64 == == d_dims: 64 == == d_mask_dims: 32 == == masked_training: True == == eyes_mouth_prio: True == == uniform_yaw: True == == blur_out_mask: True == == adabelief: True == == lr_dropout: y == == random_warp: False == == random_hsv_power: 0.1 == == true_face_power: 0.0 == == face_style_power: 0.0 == == bg_style_power: 0.0 == == ct_mode: none == == clipgrad: False == == pretrain: False == == autobackup_hour: 0 == == write_preview_history: False == == target_iter: 0 == == random_src_flip: False == == random_dst_flip: True == == batch_size: 32 == == gan_power: 0.1 == == gan_patch_size: 28 == == gan_dims: 32 == == == ==------------------ Running On ------------------== == == == Device index: 0 == == Name: NVIDIA GeForce RTX 4090 == == VRAM: 21.70GB == == ==