Are there any advanced packages that allows automated tuning of hyperparameters for neural network and traditional machine learning algorithms like XGBoost, random forest (using method like Bayesian, random search etc. that could allow faster discovering of best parameters)? I heard of hyperopt but seems there are some problems, and not sure it can train traditional machine learning algorithms?

  • $\begingroup$ I've briefly attended the Auto ML workshop last fall where automated tuning of such models was one of the topics. Especially the SMAC project seems relevant here $\endgroup$ – deemel Jan 26 '18 at 7:13

In R: mlr, mlrMBO, mlrHyperopt, irace
In Python: Spearmint and Hyperopt
Outside of this: SMAC

I would recommend mlrHyperopt, if you are a beginner and mlr/mlrMBO if you are advanced. I am a biased R user, so I do not know about the Python solutions.

| cite | improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.