init

app.py

@@ -126,7 +126,7 @@ def get_depth_image(
         feature_extractor: AutoImageProcessor,
         depth_estimator: AutoModelForDepthEstimation
 ) -> Image:
-    image_to_depth = feature_extractor(images=image, return_tensors="pt").to(device)
+    image_to_depth = feature_extractor(images=image, return_tensors="pt")#.to(device)
     with torch.no_grad():
         depth_map = depth_estimator(**image_to_depth).predicted_depth
 
@@ -180,7 +180,6 @@ class ControlNetDepthDesignModelMulti:
     """ Produces random noise images """
     def __init__(self):
         """ Initialize your model(s) here """
-        device = torch.device("cuda")
         #os.environ['HF_HUB_OFFLINE'] = "True"
         controlnet_depth= ControlNetModel.from_pretrained(
             "controlnet_depth", torch_dtype=dtype, use_safetensors=True)
@@ -199,9 +198,7 @@ class ControlNetDepthDesignModelMulti:
                                   weight_name="ip-adapter_sd15.bin")
         self.pipe.set_ip_adapter_scale(0.4)
         self.pipe.scheduler = UniPCMultistepScheduler.from_config(self.pipe.scheduler.config)
-        print(self.pipe.device)
         self.pipe = self.pipe.to(device)
-        print(self.pipe.device)
         self.guide_pipe = StableDiffusionXLPipeline.from_pretrained("segmind/SSD-1B",
                                                                     torch_dtype=dtype, use_safetensors=True, variant="fp16")
         self.guide_pipe = self.guide_pipe.to(device)
@@ -215,16 +212,6 @@ class ControlNetDepthDesignModelMulti:
         self.depth_feature_extractor, self.depth_estimator = get_depth_pipeline()
         self.depth_estimator = self.depth_estimator.to(device)
         
-        if torch.cuda.is_available():
-            # Print the number of available GPUs
-            print("Available GPU devices:")
-            for i in range(torch.cuda.device_count()):
-                print(f"Device {i}: {torch.cuda.get_device_name(i)}")
-        else:
-            print("No GPU devices available. Using CPU.")
-    
-        print(self.depth_estimator.device)
-        print(self.pipe.device)
         
     @spaces.GPU
     def generate_design(self, empty_room_image: Image, prompt: str, guidance_scale: int = 10, num_steps: int = 50, strength: float =0.9, img_size: int = 640) -> Image: