How do I create a gating function for a CNN-based object detection model
I was following this tutorial (https://tree.rocks/a-simple-way-to-understand-and-implement-object-detection-from-scratch-by-pure-cnn-36cc28143ca8) and tried to implement a gating function for my own CNN. My goal is to classify if an image has smoke, then to draw a bounding box. If there is no smoke, no bounding box should be drawn. Without the gaiting function, my code works perfectly, but it draws a box even if there is no smoke: