max_depth=None setting overridden in components._trees_hp_space about hyperopt-sklearn HOT 4 CLOSED

Thank you for your issue! I have implemented this in #183 and will close this when it gets merged.

from hyperopt-sklearn.

Nice! That was a quick response and nice work with #183.

The above described situation happens also in more general, whenever hpsklearn's value None coincides with some optional value of corresponding scikit function.
E.g. in components._trees_hp_space you have random_state=None but this is also possible value or same argument in sklearn.ensemble.RandomForestRegressor.
Second example is in your new rewrite_for_sklearn1.0.0 branch in function components.ensemble.forest._forest_hp_space the argument max_leaf_nodes=None, since it coincides with the default parameter of Random Forest function in scikit.
So there may be more possible improvements of this kind.

from hyperopt-sklearn.

Fixing the random_state parameter makes much sense in most applications.

1. It ensures that the splits that are generated are reproducible.
2. Most importantly, it ensures that the improvements are actual improvements and not improvements by pure chance.
   For a more in-depth explanation on this I would suggest checking out this towardsdatascience article by Ler Wei Han.

I will take a look at the ensemble optimizers and implement your solution in the next few days. Thanks again for raising concern on this.

from hyperopt-sklearn.

Now that it's merged I'll go ahead and close this.
Thank you for raising an issue and feel free to let us know if you find any bugs or other issues.

from hyperopt-sklearn.