-
Notifications
You must be signed in to change notification settings - Fork 3
/
train_inversion.sh
61 lines (57 loc) · 1.93 KB
/
train_inversion.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
################E4E training############################################
python encoder_inversion/train.py \
--outdir=training-runs/encoder_inversion/e4e \
--cfg=ffhq \
--data /data/zhaoxiaochen/Dataset/FFHQ/images512x512 \
--rdata /data/zhaoxiaochen/Dataset/FFHQ/orthRender256x256_face \
--gpus=8 \
--batch=32 \
--mbstd-group=4 \
--gamma=1 \
--snap=10 \
--gen_pose_cond=1 \
--model_version v20 \
--configs_path=encoder_inversion/config/train_e4e_real.yaml \
--gen_lms_cond 0 \
--training_state e4e \
--gen_mask_cond 0
################Few-shot Training###################################
python encoder_inversion/train.py \
--outdir=training-runs/encoder_inversion/few-shot \ #VFHQ/gru/both_TriNetSFT/multiT_onlyIreal \
--cfg=ffhq \
--data /data/zhaoxiaochen/Dataset/VFHQ/dataset/images512x512 \
--rdata /data/zhaoxiaochen/Dataset/VFHQ/dataset/orthRender256x256_face_eye \
--dataset_class_name encoder_inversion.dataset_video.VideoFolderDataset \
--gpus=8 \
--batch=8 \
--mbstd-group=1 \
--gamma=1 \
--tick=1 \
--snap=4 \
--gen_pose_cond=1 \
--model_version v20 \
--configs_path=encoder_inversion/config/train_textureUnet_video.yaml \
--gen_lms_cond 0 \
--gen_mask_cond 1 \
--gen_uv_cond 1 \
--training_state fewshot \
--resume training-runs/encoder_inversion/e4e/path-to-your-pkl.pkl
################Improved One-shot Training###################################
python encoder_inversion/train.py \
--outdir=training-runs/encoder_inversion/v20_128/both_TriSFT_TriplaneTexNet_uvFInp/BothSegFormerDecoder \
--cfg=ffhq \
--data /data/zhaoxiaochen/Dataset/FFHQ/images512x512 \
--rdata /data/zhaoxiaochen/Dataset/FFHQ/orthRender256x256_face_eye \
--gpus=8 \
--batch=16 \
--mbstd-group=2 \
--gamma=8 \
--snap=10 \
--gen_pose_cond=1 \
--model_version v20 \
--configs_path=encoder_inversion/config/train_textureUnet_real.yaml \
--gen_lms_cond 0 \
--gen_mask_cond 0 \
--gen_uv_cond 1 \
--training_state oneshot \
--resume training-runs/encoder_inversion/e4e/path-to-your-pkl.pkl