-
Notifications
You must be signed in to change notification settings - Fork 0
/
P1_B_inference.py
72 lines (61 loc) · 1.81 KB
/
P1_B_inference.py
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
62
63
64
65
66
67
68
69
70
71
72
import argparse
import pathlib
import random
import torch
from torchvision import transforms
from torchvision.utils import save_image
from P1_B_model import DCGAN_G
# python3 P1_B_inference.py -w ./P1_B_ckpt/best_G.pth
# pytorch-fid ./P1_B_out/ ./hw2_data/face/val/
# python3 face_recog.py --image_dir ./P1_B_out
def main(device, weight, out_dir):
# FID = 25.9896166834331, ACC = 91.100%
seed = 10
random.seed(seed)
torch.manual_seed(seed)
mean = [0.5, 0.5, 0.5]
std = [0.5, 0.5, 0.5]
UnNormalize = transforms.Normalize(
mean=[-u / s for u, s in zip(mean, std)],
std=[1 / s for s in std],
)
batch_size = 100
model = DCGAN_G().to(device)
model.load_state_dict(torch.load(weight, map_location=device))
idx = 0
for _ in range(1000 // batch_size):
z = torch.randn(batch_size, 100, 1, 1, device=device)
with torch.no_grad():
gen_imgs = model(z)
gen_imgs = UnNormalize(gen_imgs)
for img in gen_imgs:
save_image(img, out_dir / f'{idx}.png')
idx += 1
def parse_args():
parser = argparse.ArgumentParser()
parser.add_argument(
"-w", "--weight",
help="path to generator weight",
type=pathlib.Path,
default='./P1_B_ckpt/best_G.pth'
)
parser.add_argument(
"-o", "--out_dir",
help="output directory",
type=pathlib.Path,
default='./P1_B_out',
)
parser.add_argument(
"-d", "--device",
help="device",
type=torch.device,
default='cuda' if torch.cuda.is_available() else 'cpu',
)
return parser.parse_args()
if __name__ == '__main__':
args = parse_args()
try:
args.out_dir.mkdir(exist_ok=True, parents=True)
except:
pass
main(args.device, args.weight, args.out_dir)