Marius Lindauer

Results 19 issues of Marius Lindauer

**Update details:** Hi, Thanks for listing SMAC3 here. However, I'm surprised to see SMAC3 listed as BSD 1-clause, although it is BSD 3-clause. Could you please look into this again?...

update-project

* bilog (log transformations above 0 and below 0) * Gaussian Copula (ECDF -> quantiles -> Inverse Gaussian CDF)

feature

If SMAC3 tried all possible configurations in the ConfigSpace (and the target algorithm is deterministic), but the configuration budget is not yet exhausted (e.g., number of function evaluations), SMAC3 will...

bug

We round up the cutoff to the next integer so far. That's maybe necessary for the pynisher but in all other use cases (where we don't limit the runtime within...

bug

As an todo item for the next smacathon: Improve the documentation for the parallel mode. We have right now: 1. pSMAC (part of the documentation, but no example scripts), which...

documentation

By running the following code, the memory usage goes up like crazy: ``` import ConfigSpace.hyperparameters as CSH import numpy as np rnd = np.random.RandomState(19937) a = CSH.UniformIntegerHyperparameter('a', lower=1, upper=2147483647, log=True)...

bug

The yaml file to modify the pcs is fine for now, but it would be more convenient to pass a pcs file which updates the current pcs.

enhancement

Hi, As mentioned today in the meeting, the LPI plots for hyperparameters sampled on a log-scale where not on a log-scale. Cheers, Marius

Instead of a html report, we could also generate a PDF paper template output. ---> low prio

enhancement

Please train the MDS mapping only on randomly sampled configurations (not on the ones optimized wrt EI). Of course, please still plot all configurations. (As a reminder of what we...