Tesla T4 跑SAEHD模型，为啥那么2000多ms迭代一次？

IntangibleW · 发表于 2022-11-13 13:48:59

星级打分

1
2
3
4
5

平均分:NAN 参与人数:0 我的评分:未评

模型参数：

==--------------- Model Options ---------------==
==                                                       ==
==          resolution: 224                      ==
==          face_type: wf                      ==
==    models_opt_on_gpu: True             ==
==                archi: df-d                         ==
==             ae_dims: 258                      ==
==             e_dims: 66                         ==
==             d_dims: 64                         ==
==          d_mask_dims: 16                   ==
==    masked_training: True                   ==
==    eyes_mouth_prio: True                ==
==          uniform_yaw: False                   ==
==       blur_out_mask: False                ==
==          adabelief: True                      ==
==          lr_dropout: n                         ==
==          random_warp: True                ==
==    random_hsv_power: 0.0                ==
==    true_face_power: 0.01                ==
==    face_style_power: 0.0                   ==
==       bg_style_power: 0.0                   ==
==             ct_mode: lct                      ==
==             clipgrad: True                      ==
==             pretrain: False                      ==
==    autobackup_hour: 1                   ==
== write_preview_history: False             ==
==          target_iter: 0                         ==
==    random_src_flip: False                ==
==    random_dst_flip: False                ==
==          batch_size: 32                      ==
==          gan_power: 0.0                   ==
==       gan_patch_size: 24                   ==
==             gan_dims: 16                      ==
==---------------- Running On -----------------==
==                                                       ==
==       Device index: 0                      ==
==                Name: Tesla T4                ==
==                VRAM: 13.61GB             ==
==                                                       ==
=================================================

用的是猫之汉化提供的模型