add model

model.py

@@ -5,25 +5,38 @@ import torch.nn as nn
 import os
 import clip
 
-
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+model, preprocess = clip.load("ViT-B/32", device=device)
 # class GanAttack(nn.Module):
     
 class GanAttack(nn.Module):
-    def __init__(self, kernel, factor=2):
+    def __init__(self, stylegan_generator, inverter,images_resize,prompt):
         super().__init__()
 
-        self.factor = factor
-        kernel = make_kernel(kernel) * (factor ** 2)
-        self.register_buffer('kernel', kernel)
+        self.generator = stylegan_generator
+        self.generator.eval()
+        self.inverter=inverter
+        self.images_resize=images_resize
+        text=clip.tokenize(prompt).to(device)
+        with torch.no_grad():
+            self.prompt = model.encode_text(text)
+        text_len=self.prompt.shape[0]
+        self.mlp=nn.Sequential(
+            nn.Linear(text_len+512, 4096),
+                                 nn.ReLU(inplace=True),
+                                 nn.Linear(4096, 512)
+        )
 
-        p = kernel.shape[0] - factor
 
-        pad0 = (p + 1) // 2 + factor - 1
-        pad1 = p // 2
+    def forward(self, img,img_path):
+        _, _, _, refine_images, latent_codes, _=self.inverter(img,self.images_resize,img_path)
+        x=latent_codes
+        batch_size=img.shape[0]
+        prompt=self.prompt.repeat(batch_size).to(device)
+        x_prompt=torch.cat([latent_codes,prompt],dim=1)
+        x_prompt=self.mlp(x_prompt)
+        x=x_prompt+x
+        im,_=self.generator(x,input_is_latent=True, randomize_noise=False)
 
-        self.pad = (pad0, pad1)
 
-    def forward(self, input):
-        out = upfirdn2d(input, self.kernel, up=self.factor, down=1, pad=self.pad)
-
-        return out
+        return img,refine_images,im,x