Sebastian Raschka

In some cases that might happen if there are things that are very quick to calculate and there is an overhead of spawning subprocess. In most cases though, for longer...

A bit of a background story, the `VotingClassifier` in scikit-learn actually comes from the one in mlxtend. I had a very rudimentary version in my Python Machine Learning book in...

Regarding this PR, I just see that there is a file mlxtend/regressor/tests/test_ensemble_voting_regressor.ipynb However, the test file should be a regular .py script similar to https://github.com/rasbt/mlxtend/blob/master/mlxtend/classifier/tests/test_ensemble_vote_classifier.py The jupyter documentation ensemble_voting_regressor.ipynb should...

No worries! Please do not make a new PR though. You can just update this existing PR. After making the changes, simply `git add` and `git commit` these changes and...

I agree, having additional metrics would be nice.

Hm, good point. But are the outputs of the `decision_function` on the same scale (or [0, 1] range) like probabilities? I think that SVC, for example, computes the distance from...

I was thinking of the EnsembleVoteClassifier where the probabilities can be averaged via soft-voting, which would be a problem if the `decision_function` values were on a different scale than the...

> What if the meta_clf_ has no predict_proba ... In that case, when calling my `stackingclassier.predict_proba`, it should raise an error. Alternatively, this method (`stackingclassier.predict_proba`) could be removed via the...

I like this overall idea. There is some rudimentary DataFrame support now (http://rasbt.github.io/mlxtend/user_guide/feature_selection/SequentialFeatureSelector/#example-11-using-pandas-dataframes) but yeah, I think it converts to NumPy arrays internally. So with your `DataFrameWrapper`, the data frame...

Hm, this sounds interesting but tricky. You can actually pass an axes object to the function, which you would need for a 3D projection, i.e., sth like ax = fig.add_subplot(111,...