Shijie. Li
Shijie. Li
pip install mmcv=0.2.16 may solve
> I train the source code on 07train+12train sets of PASCAL VOC, and test on 07test set. The mAP is **66.13**. Training loss is around **1** while test loss between...
the compute with map in eval should be mAP/=(len(eval_dataset.CLASSES_NAME)-1) but not +1?
> Have you solved the problem? I got the same error.@KevinQian97 @WangTianYuan I found that if you use your own dataset to train the model, if it has dirty data,...