Hello! It appears there are some significant gaps in the provided code. There are various instance variables missing from the initialization methods. While the missing code can generally be inferred, the following code on line 127 of ddpg.py presents a problem. The _critic object does not have a method train. What is intended? Or, what is the train method for the critic?
self._critic.train(states, actions, q_targets)
Hello! It appears there are some significant gaps in the provided code. There are various instance variables missing from the initialization methods. While the missing code can generally be inferred, the following code on line 127 of ddpg.py presents a problem. The
_criticobject does not have a methodtrain. What is intended? Or, what is thetrainmethod for the critic?self._critic.train(states, actions, q_targets)