epoches: 329
train set - pack1 : 700
train set - pack2(after augmented data) : 1085
test set : 465
Train set - pack2 weights : mean average precision (mAP@0.50) = 0.974545, or 97.45 %
Test set - pack1 weights : mean average precision (mAP@0.50) = 0.944054, or 94.41 %
Test set - pack2 weights : mean average precision (mAP@0.50) = 0.974545, or 97.45 %
Also i trained my dataset on YOLOv5s with 100 epoches, and its my result:
