SongyiGao

> >  > > 我想知道“教师强制：将目标label作为下一个输入”是在干什么？ > > 加速模型收敛的过程 使用教师强制和不用教师强制训练出来的最终模型精度是否会不一样呢？

> > For efficient update. > > Weight normalization reparameterizes the gradient into direction and length. > > And for baseline++, only the direction of weight is important. > >...

Can you provide a record chart of other indicators during training to help locate the problem?

Thank you for your suggestion. The current offline dataset has been collected based on a version prior to V2, so the environments may vary between different versions, which could lead...

> 您好，请问flow 代指什么安装包呢，我使用pip install flow 它告诉我没有这个包 您好，强化学习环境中flow通常是指交通流量模拟库 "Flow"，它是一个用于模拟交通流量和进行交通管理研究的开源库。具体请参考：https://flow.readthedocs.io/en/latest/index.html