Hi,
I have a doubt about the code used in generating the model output. Do we get directly xmin, ymin, xmax, ymax from the model output?
Line 497 in the train.py mentions the output obtained by the model when it's used in eval mode as xmin, ymin, xmax, ymax.
Usually, the model gives x_center, y_center, width, height as output in Yolo and needs to be converted to xmin,ymin, xmax, ymax. Can you clarify the output format produced by the current implementation. Is it xmin, ymin, xmax, and ymax?
Hi, I have a doubt about the code used in generating the model output. Do we get directly xmin, ymin, xmax, ymax from the model output?
Line 497 in the train.py mentions the output obtained by the model when it's used in eval mode as xmin, ymin, xmax, ymax. Usually, the model gives x_center, y_center, width, height as output in Yolo and needs to be converted to xmin,ymin, xmax, ymax. Can you clarify the output format produced by the current implementation. Is it xmin, ymin, xmax, and ymax?