Run one timestep of the environment's dynamics.
(self, state: State, action: jax.Array)
| 38 | return State(pipeline_state, obs, reward, done, metrics) |
| 39 | |
| 40 | def step(self, state: State, action: jax.Array) -> State: |
| 41 | """Run one timestep of the environment's dynamics.""" |
| 42 | pipeline_state = self.pipeline_step(state.pipeline_state, action) |
| 43 | obs = self._get_obs(pipeline_state) |
| 44 | reward = jp.cos(pipeline_state.q[1]) - jp.abs(pipeline_state.qd[0]) |
| 45 | done = 0.0 |
| 46 | return state.replace( |
| 47 | pipeline_state=pipeline_state, obs=obs, reward=reward, done=done |
| 48 | ) |
| 49 | |
| 50 | @property |
| 51 | def action_size(self): |