lzh-awesome

hi,i am trying to reproduce the battle game (red army use MFQ and blue army use DQN.) i trained the model and find both algorithums has a negative reward.and the...

 问题的初衷： 我想要把用MFAC训练好的模型和用AC训练好的模型进行对比，让这两个算法进行交叉战斗实验。在终端运行如下：python battle.py --algo mfac --oppo ac --idx 10 10 仍然抛出异常： ValueError: The passed save_path is not a valid checkpoint: /home/lzh/postgraduate project/mfrl-master/data/models/mfac-0/mfac_1 疑惑的点： 1）如何运行battle.py 2) 参数--idx 如何设置parser.add_argument('--idx', nargs='*', required=True)...