Why don't you just use qval?

[Triple-ROCK]

gym/gym/envs/mujoco/half_cheetah.py

Line 16 in 6df1b99

reward_run = (xposafter - xposbefore)/self.dt

[Triple-ROCK]

gym/gym/envs/mujoco/half_cheetah.py

Line 16 in 6df1b99

reward_run = (xposafter - xposbefore)/self.dt

I see some tutorial where they register customized env and use 'xvel = observations[:, 9]' instead of
reward_run = (xposafter - xposbefore)/self.dt

[jkterry1]

PR #2762 is about to be merged, introducing V4 MuJoCo environments using new bindings and a dramatically newer version of the engine. If this issue still persists with the V4 ones, please create a new issue for it.