Can I use the metrics.py code to evaluate on images that do not contain objects?

[A00TB]

Hi,

I just want to make sure I haven't caused any confusion here. I am evaluating on images contain both objects and images that contain no objects. I'm wondering if this will cause issues later on, for example the true_Positives list will not contain detections that have no objects, and thus not contain those images, whereas the false_Positives list will potentially contain predicted objects in an image that has no TP image...