"RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.FloatTensor [64, 1]], which is output 0 of AsStridedBackward0, is at version 3; expected version 2 instead. Hint: the backtrace further a

Hello @Sangyoon1207

I have been faced with the same error in a MultiAgent environment. I am currently looking for a solution but I strongly suspect it would be something related to this post. Hopefully you find it helpful