AttGAN Usage
-
training
-
for 128x128 images
CUDA_VISIBLE_DEVICES=0 \ python train.py \ --load_size 143 \ --crop_size 128 \ --model model_128 \ --experiment_name AttGAN_128
-
for 128x128 images with all 40 attributes!
CUDA_VISIBLE_DEVICES=0 \ python train.py \ --load_size 143 \ --crop_size 128 \ --model model_128 \ --experiment_name AttGAN_128_40 \ --att_names 5_o_Clock_Shadow Arched_Eyebrows Attractive Bags_Under_Eyes \ Bald Bangs Big_Lips Big_Nose Black_Hair Blond_Hair Blurry \ Brown_Hair Bushy_Eyebrows Chubby Double_Chin Eyeglasses Goatee \ Gray_Hair Heavy_Makeup High_Cheekbones Male Mouth_Slightly_Open \ Mustache Narrow_Eyes No_Beard Oval_Face Pale_Skin Pointy_Nose \ Receding_Hairline Rosy_Cheeks Sideburns Smiling Straight_Hair \ Wavy_Hair Wearing_Earrings Wearing_Hat Wearing_Lipstick \ Wearing_Necklace Wearing_Necktie Young
-
for 256x256 images
CUDA_VISIBLE_DEVICES=0 \ python train.py \ --load_size 286 \ --crop_size 256 \ --model model_256 \ --experiment_name AttGAN_256
-
for 384x384 images
CUDA_VISIBLE_DEVICES=0 \ python train.py \ --load_size 429 \ --crop_size 384 \ --model model_384 \ --experiment_name AttGAN_384
-