...
Assertion: Given an image I, if I has pylons in it assert that the result of the predict function on I has a non-zero number of boxes, else assert that it has no bounding boxes. Manually assert the position of the bounding boxes. Automating the check for position of bounding boxes can also be done by taking the average centoid of all bounding boxes in I and ensuring it is within 10% after 2 runs, in a form of regression testing. See the comment for more details.
Expected Data: In this case the test developer will manually label each image file has having or not having bounding boxes.
...