mlrHyperopt
mlrHyperopt copied to clipboard
jakob-r
Easy Hyper Parameter Optimization with mlr and mlrMBO.
When using mlrHyperopt very big objects get saved. This can be very annoying when doing repeated cross validation on several datasets, cause big objects are saved/used. Example: ``` task =...
Can you include the possibility of priors instead of only ranges to your package? Would be awesome, although I think that it is not easy to implement this in mlrMBO......
Maybe I have overseen something, but I get different optimal results in the plot and with getTuneResult. I really like the plotOptPath function! ``` par.set = makeParamSet( makeNumericParam("C", lower =...
Based on this guide (https://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf), I uploaded new parameter settings for svm. Maybe you should make them the defaults.
Hi Jakob, Is it possible to update an already uploaded par.config? I do not want to spam with a lot of different par.configs. I did a test [upload](http://mlrhyperopt.jakob-r.de/parconfigs/54) (ID: 54)...
I looked up your default parameter space for ranger and this is not really suitable for bigger datasets. ``` Type len Def Constr Req Tunable Trafo mtry integer - 2...
In a `HyperControl` object, is there a way to change either `measure` or `resampling` (or both) and leave other settings at their defaults? This code results in error: ```r makeHyperControl(resampling...
