Update app.py

app.py

@@ -35,7 +35,7 @@ image_encoder = CLIPVisionModelWithProjection.from_pretrained(
 ).to(dtype=torch.float16, device=device)
 
 ip_img_size = 336
-clip_image_processor = CLIPImageProcessor(size=ip_img_size, crop_size=ip_img_size).to(device)
+clip_image_processor = CLIPImageProcessor(size=ip_img_size, crop_size=ip_img_size)
 
 pipe = StableDiffusionXLPipeline(
     vae=vae,
@@ -64,7 +64,7 @@ def infer(prompt, ip_adapter_image, negative_prompt="", seed=42, randomize_seed=
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
     
-    generator = torch.Generator(device="cpu").manual_seed(seed)
+    generator = torch.Generator(device="cuda").manual_seed(seed)
     
     pipe.set_ip_adapter_scale([ip_adapter_scale])