Update pipeline.py

Specify the device for the `input_image_tensor` to avoid encountering the following error code when employing GPU for image-to-image operations.

```
RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same or input should be a MKLDNN tensor and weight is a dense tensor
```

sd/pipeline.py

@@ -88,7 +88,7 @@ def generate(
             # (Height, Width, Channel)
             input_image_tensor = np.array(input_image_tensor)
             # (Height, Width, Channel) -> (Height, Width, Channel)
-            input_image_tensor = torch.tensor(input_image_tensor, dtype=torch.float32)
+            input_image_tensor = torch.tensor(input_image_tensor, dtype=torch.float32, device=device)
             # (Height, Width, Channel) -> (Height, Width, Channel)
             input_image_tensor = rescale(input_image_tensor, (0, 255), (-1, 1))
             # (Height, Width, Channel) -> (Batch_Size, Height, Width, Channel)
@@ -167,4 +167,4 @@ def get_time_embedding(timestep):
     # Shape: (1, 160)
     x = torch.tensor([timestep], dtype=torch.float32)[:, None] * freqs[None]
     # Shape: (1, 160 * 2)
-    return torch.cat([torch.cos(x), torch.sin(x)], dim=-1)
+    return torch.cat([torch.cos(x), torch.sin(x)], dim=-1)