Fix the order of the input tensors

The order of the input tensors is incorrect and therefore the sample notebook does not run.

The sample code has the order of the input tensors as `boxes, classes, scores, num_detections`, but actually the order should be `scores, boxes,  num_detections, classes`

After correcting the indexes of the input tensor, the sample notebook runs.

research/object_detection/colab_tutorials/centernet_on_device.ipynb

@@ -244,10 +244,10 @@
         "\r\n",
         "  interpreter.invoke()\r\n",
         "\r\n",
-        "  boxes = interpreter.get_tensor(output_details[0]['index'])\r\n",
-        "  classes = interpreter.get_tensor(output_details[1]['index'])\r\n",
-        "  scores = interpreter.get_tensor(output_details[2]['index'])\r\n",
-        "  num_detections = interpreter.get_tensor(output_details[3]['index'])\r\n",
+        "  scores = interpreter.get_tensor(output_details[0]['index'])\r\n",
+        "  boxes = interpreter.get_tensor(output_details[1]['index'])\r\n",
+        "  num_detections = interpreter.get_tensor(output_details[2]['index'])\r\n",
+        "  classes = interpreter.get_tensor(output_details[3]['index'])\r\n",
         "\r\n",
         "  if include_keypoint:\r\n",
         "    kpts = interpreter.get_tensor(output_details[4]['index'])\r\n",
@@ -759,4 +759,4 @@
       ]
     }
   ]
-}
+}
\ No newline at end of file