Spaces:
Runtime error
Runtime error
Ahsen Khaliq
commited on
Commit
•
c3d2c2f
1
Parent(s):
49ce528
add yasuho model
Browse files
app.py
CHANGED
@@ -52,6 +52,9 @@ generatorjinx = deepcopy(original_generator)
|
|
52 |
|
53 |
generatorcaitlyn = deepcopy(original_generator)
|
54 |
|
|
|
|
|
|
|
55 |
|
56 |
|
57 |
transform = transforms.Compose(
|
@@ -85,6 +88,11 @@ os.system("gdown https://drive.google.com/uc?id=1cUTyjU-q98P75a8THCaO545RTwpVV-a
|
|
85 |
ckptcaitlyn = torch.load('arcane_caitlyn_preserve_color.pt', map_location=lambda storage, loc: storage)
|
86 |
generatorcaitlyn.load_state_dict(ckptcaitlyn["g"], strict=False)
|
87 |
|
|
|
|
|
|
|
|
|
|
|
88 |
|
89 |
def inference(img, model):
|
90 |
aligned_face = align_face(img)
|
@@ -99,9 +107,12 @@ def inference(img, model):
|
|
99 |
elif model == 'Jinx':
|
100 |
with torch.no_grad():
|
101 |
my_sample = generatorjinx(my_w, input_is_latent=True)
|
102 |
-
|
103 |
with torch.no_grad():
|
104 |
my_sample = generatorcaitlyn(my_w, input_is_latent=True)
|
|
|
|
|
|
|
105 |
|
106 |
|
107 |
npimage = my_sample[0].permute(1, 2, 0).detach().numpy()
|
|
|
52 |
|
53 |
generatorcaitlyn = deepcopy(original_generator)
|
54 |
|
55 |
+
generatoryasuho = deepcopy(original_generator)
|
56 |
+
|
57 |
+
|
58 |
|
59 |
|
60 |
transform = transforms.Compose(
|
|
|
88 |
ckptcaitlyn = torch.load('arcane_caitlyn_preserve_color.pt', map_location=lambda storage, loc: storage)
|
89 |
generatorcaitlyn.load_state_dict(ckptcaitlyn["g"], strict=False)
|
90 |
|
91 |
+
os.system("gdown https://drive.google.com/uc?id=1SKBu1h0iRNyeKBnya_3BBmLr4pkPeg_L")
|
92 |
+
|
93 |
+
ckptyasuho = torch.load('jojo_yasuho_preserve_color.pt', map_location=lambda storage, loc: storage)
|
94 |
+
generatoryasuho.load_state_dict(ckptyasuho["g"], strict=False)
|
95 |
+
|
96 |
|
97 |
def inference(img, model):
|
98 |
aligned_face = align_face(img)
|
|
|
107 |
elif model == 'Jinx':
|
108 |
with torch.no_grad():
|
109 |
my_sample = generatorjinx(my_w, input_is_latent=True)
|
110 |
+
elif model == 'Caitlyn':
|
111 |
with torch.no_grad():
|
112 |
my_sample = generatorcaitlyn(my_w, input_is_latent=True)
|
113 |
+
else:
|
114 |
+
with torch.no_grad():
|
115 |
+
my_sample = generatoryasuho(my_w, input_is_latent=True)
|
116 |
|
117 |
|
118 |
npimage = my_sample[0].permute(1, 2, 0).detach().numpy()
|