有关RTMDet的backbone&neck参数配比的问题(Questions about the ratio of the backbone&neck parameters of RTMDet) #935
Unanswered
little-spoon
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
中文:RTMDet里面提到的 backbone {expand_ratio} 和 neck {expand_ratio} 这两个参数的和是为1的吗?那为什么论文中Table5(c)中两个值的和不为1呢?这两个值的具体含义究竟是什么呢?我如果进行不同配比的实验,具体应该如何调整这两个值才是比较合理的呢?(例如0.6/0.4,0.4/0.6,04/0.4,0.5/0.4等等,这些是否都是合理的配比?)
English:Is the sum of backbone {expand_ratio} and neck {expand_ratio} mentioned in RTMDet equal to 1? So why doesn't the sum of the two values in Table5(c) in the paper equal 1? What exactly do these two values mean? If I conduct experiments with different ratios, how should I adjust these two values is more reasonable? (e.g. 0.6/0.4, 0.4/0.6, 04/0.4, 0.5/0.4, etc., are these all reasonable ratios?)
Beta Was this translation helpful? Give feedback.
All reactions