WongKinYiu / yolov7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
GNU General Public License v3.0
13.4k stars 4.23k forks source link

onnx non end2end #2057

Open thestars-maker opened 4 months ago

thestars-maker commented 4 months ago

In the model export using export.py export, don't use end2end,onnx the output dimension is float32[1,25200,85],, may I ask this 85 is such as refers to those dimensions it [boxes,prob,per classes conf], this boxes refers to x,y,w,h or x1 y1,x2,y2?