search

AlexeyAB / darknet

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )
http://pjreddie.com/darknet/
Other
21.6k stars 7.95k forks source link

Getting Average Recall #5295

Open Linch727 opened 4 years ago

Linch727 commented 4 years ago

Hi,

I would like to calculate the Average Recall (https://manalelaidouni.github.io/manalelaidouni.github.io/Evaluating-Object-Detection-Models-Guide-to-Performance-Metrics.html#average-recall-ar). I used the commands ./darknet detector map cfg/coco.data cfg/yolov3.cfg yolov3.weights -iou_thresh 0.50 -points 101 with 11 different thresholds ([0.5:0.95]), than calculated the area under the iou-recall curve, but the result was very low, although the detection results are fantastic.

Is there another way to get the average recall?

AlexeyAB commented 4 years ago

Use additioal flag -thresh 0.001

Linch727 commented 4 years ago

Use additioal flag -thresh 0.001

Thanks for your reply. What is the function of this flag? It seems to change nothing...