MCPcopy Create free account
hub / github.com/google-deepmind/dm_control / after_step

Method after_step

dm_control/rl/control.py:301–312  ·  view source on GitHub ↗

Optional method to update the task after the physics engine has stepped. Called by `control.Environment` after stepping the physics engine and before `control.Environment` calls `get_observation, `get_reward` and `get_termination`. The default implementation is a no-op. Args:

(self, physics)

Source from the content-addressed store, hash-verified

299 """
300
301 def after_step(self, physics):
302 """Optional method to update the task after the physics engine has stepped.
303
304 Called by `control.Environment` after stepping the physics engine and before
305 `control.Environment` calls `get_observation, `get_reward` and
306 `get_termination`.
307
308 The default implementation is a no-op.
309
310 Args:
311 physics: Instance of `Physics`.
312 """
313
314 @abc.abstractmethod
315 def action_spec(self, physics):

Callers 1

stepMethod · 0.45

Calls

no outgoing calls

Tested by

no test coverage detected