Jendrik Seipp

#478 has been merged now. Does the latest RedNotebook version open correctly for you now?

I'll try that, thanks!

We changed the code to ``` scenario = Scenario({ .... "acq_opt_challengers": 1000, }) ``` but still get the same error messages. Can it be that the setting is not picked...

In smac_hpo_facade.py I found the following code snippet: ``` # better improve acquisition function optimization # 2. more randomly sampled configurations self.solver.scenario.acq_opt_challengers = 10000 ``` I think the value should...

Could it be that the problem is that no tested configuration is better than the initial incumbent?

I have reduced the run to a toy example (test-numpy.py). When I use "ulimit -Sv 600000" and then "rm -rf smac && ./test-numpy.py", I get "MemoryError: Unable to allocate array...

Thanks for your comments! Reg. 1: I agree that it would be better to not limit the virtual memory, but we have to make sure that the SMAC runs don't...

Thanks! I'll try that.

I just found out that `ulimit -m` has no effect on modern Linux: https://unix.stackexchange.com/questions/129587/does-ulimit-m-not-work-on-modern-linux BTW, setting `acq_opt_challengers = 1000` removed the numpy memory error for us (even if users shouldn't...

Thanks for the quick answer, Marius! I'm optimizing for solution quality. Here is a minimal working example. Called with `./mwe.py --random-seed=0 --modules=20 --overall-time-limit 30`, it uses 20*9 hyperparameters and checks...