|
星级打分
平均分:NAN 参与人数:0 我的评分:未评
本帖最后由 aixx 于 2023-9-10 10:48 编辑
用底丹,发现都没调参的项呢,直接就练了,我看教程,至少pretrain参数是可以调的呀。主要是导出dfm后,使用还是目标视频的人脸,没看有换脸呢,很不解。请问各位大佬,这是怎么回事呀?
训练过程如下:
[0] : 384 - latest
[1] : 192wf
[2] : ZIBI2.0
[3] : 猫之汉化神丹224 df-d
: 0
0
Loading 384_SAEHD model...
Choose one or several GPU idxs (separated by comma).
[CPU] : CPU
[0] : NVIDIA GeForce RTX 3090
[0] Which GPU indexes to choose? : 0
0
Initializing models: 100%|###############################################################| 5/5 [00:11<00:00, 2.22s/it]
Loading samples: 100%|############################################################| 2390/2390 [00:04<00:00, 489.93it/s]
Loading samples: 100%|############################################################| 1619/1619 [00:05<00:00, 303.70it/s]
================== Model Summary ===================
== ==
== Model name: 384_SAEHD ==
== ==
== Current iteration: 4313541 ==
== ==
==---------------- Model Options -----------------==
== ==
== resolution: 384 ==
== face_type: wf ==
== models_opt_on_gpu: True ==
== archi: df-udt ==
== ae_dims: 448 ==
== e_dims: 96 ==
== d_dims: 96 ==
== d_mask_dims: 32 ==
== masked_training: True ==
== eyes_mouth_prio: False ==
== uniform_yaw: False ==
== blur_out_mask: False ==
== adabelief: True ==
== lr_dropout: y ==
== random_warp: False ==
== random_hsv_power: 0.0 ==
== true_face_power: 0.0 ==
== face_style_power: 0.0 ==
== bg_style_power: 0.0 ==
== ct_mode: none ==
== clipgrad: True ==
== pretrain: False ==
== autobackup_hour: 6 ==
== write_preview_history: False ==
== target_iter: 0 ==
== random_src_flip: False ==
== random_dst_flip: False ==
== batch_size: 8 ==
== gan_power: 0.0 ==
== gan_patch_size: 48 ==
== gan_dims: 16 ==
== ==
==------------------ Running On ------------------==
== ==
== Device index: 0 ==
== Name: NVIDIA GeForce RTX 3090 ==
== VRAM: 21.17GB ==
== ==
====================================================
Starting. Press "Enter" to stop training and save model.
[01:11:34][#4314920][5487ms][0.1467][0.0923]
[01:36:15][#4316351][5440ms][0.1291][0.0784]
[02:01:13][#4317793][4459ms][0.1227][0.0740]
[02:26:13][#4319247][4540ms][0.1179][0.0713]
[02:51:13][#4320723][5250ms][0.1147][0.0694]
[03:16:13][#4322200][4239ms][0.1117][0.0682]
[03:41:12][#4323676][4299ms][0.1102][0.0669]
[04:06:14][#4325153][5092ms][0.1079][0.0660]
[04:31:12][#4326627][4271ms][0.1063][0.0652]
[04:56:13][#4328103][5269ms][0.1051][0.0643]
[05:21:12][#4329577][4265ms][0.1038][0.0636]
[05:46:13][#4331053][4323ms][0.1026][0.0634]
[06:11:15][#4332528][6861ms][0.1018][0.0627]
[06:36:14][#4334004][5269ms][0.1007][0.0623]
[07:01:21][#4335481][4438ms][0.0993][0.0617]
[07:26:13][#4336947][4273ms][0.0986][0.0613]
[07:51:13][#4338425][4316ms][0.0983][0.0609]
[08:16:13][#4339903][4222ms][0.0974][0.0606]
[08:41:14][#4341381][5868ms][0.0961][0.0602]
[09:06:13][#4342859][5079ms][0.0961][0.0599]
[09:31:13][#4344340][4311ms][0.0953][0.0595]
[09:48:18][#4345317][0997ms][0.0737][0.0605]
|
|