Confused with your mAP computing!

[fighting-liu]

@BobLiu20 Thanks for your code, but your mAP computing is actually recall at specific confidence threshold, which is totally different from mAP. Here is my test results on your models with coco-evaluate tools.

Input size: 416x416

YOLOv3 (convert from paper) mAP: 0.291(0.31 for paper) mAP(0.5): 0.532 (0.553 for paper)

YOLOv3 (our train 20 epochs) mAP:0.239 mAP(0.5): 0.461

[BobLiu20]

@fighting-liu Hi, which val dataset using in your testing. 5k or 40504?
I would like to try it now. thanks.

[fighting-liu]

@BobLiu20 I just use your val dataset 5k.

[wangxianrui]

@fighting-liu Hi, when I finishing training, how can I create the json file to evaluate the model, can you provide some script, thanks

[yanxp]

@fighting-liu I meet the same problem. So the true mAP is 0.239?