Hi there! In /yolov3_tf2/models.py there is division of 416 for the anchor boxes. I custom trained yolov3 with 512x512 resolution. So do I have to change the divison with 512 instead of 416?
code:
yolo_anchors = np.array([(10, 13), (16, 30), (33, 23), (30, 61), (62, 45),
(59, 119), (116, 90), (156, 198), (373, 326)],
np.float32) / 416
Edit: Also in object_tracker.py we should not forget to change the size to the detect images:
flags.DEFINE_integer('size',416, 'resize images to') #change size according to resolution.
Hi there! In /yolov3_tf2/models.py there is division of 416 for the anchor boxes. I custom trained yolov3 with 512x512 resolution. So do I have to change the divison with 512 instead of 416? code: yolo_anchors = np.array([(10, 13), (16, 30), (33, 23), (30, 61), (62, 45), (59, 119), (116, 90), (156, 198), (373, 326)], np.float32) / 416 Edit: Also in object_tracker.py we should not forget to change the size to the detect images: flags.DEFINE_integer('size',416, 'resize images to') #change size according to resolution.
Thanks.