When I trained the code on my dataset, the mAP always kept 0.0.

toandaominh1997 / EfficientDet.Pytorch

Implementation EfficientDet: Scalable and Efficient Object Detection in PyTorch

MIT License

1.44k stars 305 forks source link

When I trained the code on my dataset, the mAP always kept 0.0. #140

Open Williamwsk opened 4 years ago

Williamwsk commented 4 years ago

When I trained the code on my dataset, the mAP always kept 0.0. My dataset had 17 classes, how can I fix it? Hope to your reply. mAP: len(VOC_CLASSES): 17 A: 0.0 B: 0.0 C: 0.0 D: 0.0 T: 0.0 F: 0.0 score: 0.0 anum: 0.0 bnum: 0.0 cnum: 0.0 rnum: 0.0 onum: 0.0 title: 0.0 keyword1: 0.0 keyword2: 0.0 keyword3: 0.0 keyword4: 0.0 avg mAP: 0.0 55 epoch: start training....

Wangyy-moya commented 3 years ago

I got the same issue.hope the author reply

ZixuanLingit666 commented 3 years ago

I also have this problem, but my loss in trainning is normal

ProjectCase commented 7 months ago

I have the same problem, the mAP is always 0, but the loss is converging normally