search

facebookresearch / hanabi_SAD

Simplified Action Decoder for Deep Multi-Agent Reinforcement Learning
Other
97 stars 35 forks source link

There seems to be a bug in pyhanabi/r2d2.py #30

Closed PaladinEE15 closed 2 years ago

PaladinEE15 commented 2 years ago

pyhanabi/r2d2.py", line 69 assert priv_s.dim() == 2, "dim should be 2, [batch, dim], get %d" % s.dim()

s is not defined. Seems that it should be 'priv_s'?

hengyuan-hu commented 2 years ago

yes! Btw, there is a newer and better repo (https://github.com/facebookresearch/off-belief-learning) that implements the algorithms in this repo and more.

PaladinEE15 commented 2 years ago

Thanks for the reminder! I'll check it out