[rllib] Allow envs to be auto-registered; add on_train_result callback with curriculum example#3451
Conversation
|
Test FAILed. |
|
Test FAILed. |
| if self.config["callbacks"].get("on_train_step"): | ||
| self.config["callbacks"]["on_train_step"]({ | ||
| "agent": self, | ||
| "result": result, |
There was a problem hiding this comment.
I think the advantage here is the user can mutate it if they want.
| self.__setstate__(extra_data) | ||
|
|
||
|
|
||
| def _register_if_needed(env_object): |
There was a problem hiding this comment.
should this happen earlier? i.e., before the config is pickled, somewhere in trial.py
There was a problem hiding this comment.
I don't think it really matters, since the pickled config can't be resolved anyways. Basically it just adds a little more text there.
|
Test FAILed. |
|
Test FAILed. |
|
Test FAILed. |
|
Test FAILed. |
|
Test FAILed. |
|
Test FAILed. |
|
|
||
| .. image:: custom_metric.png | ||
|
|
||
| Example: Curriculum Learning |
There was a problem hiding this comment.
Can probably put together a page full of examples (in a later PR)
|
Test FAILed. |
ericl
added
the
tests-ok
3 participants
No description provided.