JierunChen
commented
1 year ago @WYHZQ Hi, table 6 specifies the training and evaluation settings only for ImageNet-1k classification but not for detection. And it's not unusual that the optimal value of drop_path_rate varies for different tasks.
作者您好！ 最近在研读您的论文和代码时发现论文、cfg和model文件中对fasternet_s的drop_path_rate描述有不一致的地方。 fasternet_s.yaml文件中定义drop_path_rate=0.1，这与论文中table 6的描述是一致的。但是detection/backbones/fasternet.py中对fasternet_s的drop_path_rate却定义为0.15 请问用drop_path_rate=0.1的backbone预训练权重在检测任务上定义drop_path_rate=0.15没有什么影响吗？