Sven Mika comments

Results 40 comments of


                                            Sven Mika

[RLlib] Ape-X crashes for custom environment with Dict observation space

Hey @eleninisioti , thanks for posting this issue here! I tried to reproduce it, but ran into another, seemingly related error message: ``` The observation collected from env.reset() was not...

[RLLib] Implement evaluation callbacks

Thanks for this awesome change @olipinski , and for the review @gjoliver ! :)

[RLlib] Eval workers use async req manager.

Hey @ArturNiederfahrenhorst , please take another look. All your questions have been addressed now. Thanks!

[RLlib] Fix SAC config parameter that is not used

Hey @ArturNiederfahrenhorst , could you double-check the test 2 case failures? Seems connector related, but these tests are currently passing in master, so just making 100% sure. Thanks

[RLlib] Local worker alias

Hey @ArturNiederfahrenhorst , thanks for taking the initiative here. Another simplification suggestion would be to completely remove the local worker from the WorkerSets (we do this already for evaluation workersets,...

[tune] Error saving checkpoint based on nested metric score

Thanks for your help on this, @xwjiang2010 ! :)

[RLlib] Remove unneeded args from offline learning examples

Hey @gjoliver , could you approve this, so it can be merged, now that questions have been addressed?

[RLlib] Unable to call ray.remote functions inside env/action dist

Hey @HJasperson , thanks for filing this issue. Your code looks totally fine and I wouldn't know of any reason, why this shouldn't work. I have a few questions, though....

[RLlib] Unable to call ray.remote functions inside env/action dist

There are a couple more flaw in your repro. Please make sure to test your reproduction scripts to show the reported error, before opening the issue. * `ray.get([self.slowFunction.remote(self.someProperty[i]) for i...

[Algorithms] Return dict (instead of 2-tuple) from API method: get_action.

In the pipeline, will be done in 0.6.x.