Update inf.py

inf.py

@@ -111,9 +111,9 @@ class InferencePipeline:
         if not torch.cuda.is_available():
             raise gr.Error('CUDA is not available.')
 
-        print(f"running with {self.device}, {self.pipe.device}")
-
         self.load_pipe(content_lora_model_id, style_lora_model_id, content_alpha, style_alpha)
+        
+        self.pipe.to(self.device)
 
         generator = torch.Generator().manual_seed(seed)
         out = self.pipe(