mAP is not good? #8
Description
I trained the model with my own single label datasets and model with 130 number of epotch with 1000 epotch length and i got around 82% . Also i tested the model and the confidence is between 90%-99% but the mAP is not good, which is 0.42. What should i do? Why is this happening?