This repository has been archived by the owner on Apr 6, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 38
too many anchor boxes #7
Comments
Closed
可能是训练好的都是20类的吧 |
别恢复所有层的权重,调整一下学习速率,重新训练试一下 |
我只有最后一层的weights 变量没有加载,毕竟我自己的数据集只有一个类嘛。 |
最后几层的权重都不要恢复,你再试一下。 |
请问你是从scratch开始用voc训练的吗?一开始梯度爆炸的问题是怎么解决的? |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
大神你好,
我加载了你提供的VOC数据集的weight继续训练我自己的数据集(只有一个类:house)。
进行到3000+ steps的时候loss突然变成nan,此前在1.0上下波动。
我加载了最新的ckpt,test结果图片上遍布密集的anchor box,如图:
想请教大佬一下大概是哪里有问题,请问您提供的weight也是通过这个项目训练出来的吗?
The text was updated successfully, but these errors were encountered: