-
Notifications
You must be signed in to change notification settings - Fork 34
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
模型训练的问题 #17
Comments
哪个环境呢?这种情况之前有的,说明其实reward设计还不够strong,所以在policy更新的时候,一旦跑偏就不太好纠正回来。还有一种可能就是学习率太大了,一次更新太多破坏了原来的policy网络,然后就又相当于重新开始了。 |
谢谢,那大概率是奖励设置的问题,我再调调参。环境是自己建的二维场景,就几根柱子。我想让环境从易到难,逐步训练模型,否则直接在复杂环境中训练始终无法收敛。😞 |
按照描述这种场景用no_cnn应该会挺快收敛的,可以看一下主要失败原因是碰撞还是到达目标之后转圈,如果是转圈的话可以适当增加目标位置的可接受半径 |
小障碍物应该是特征的问题,只要下采样之后还能够看到障碍物即可,讲道理都是可以避开的,你可以研究一下下采样对于小障碍物的影响情况。 关于加速的问题,我也没办法跑满,具体原因没有去细查- -! |
好的 太小的障碍物下采样肯定就丢了 比如AirSimNH里面的电线或者细杆子,经常撞上去😂 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
请问遇到过下图的问题吗?模型训练的过程中成功率先逐渐上升到0.95左右,后面接着训练成功率又掉下去了。算法是SAC,神经网络是NoCNN,训练过程中没有修改任何参数。
The text was updated successfully, but these errors were encountered: