Johannes Dornheim

Hi, it seems like the state features that are added to the replay buffer in the HindsightExperienceReplayWrapper are features for the full observation-dictionary (observation, achieved-goal, desired-goal). https://github.com/hill-a/stable-baselines/blob/cfcdb2fc79260b25aa0c32faf4cf5d551f166f29/stable_baselines/her/replay_buffer.py#L153 https://github.com/hill-a/stable-baselines/blob/cfcdb2fc79260b25aa0c32faf4cf5d551f166f29/stable_baselines/her/replay_buffer.py#L184 the feature-vectors...