CODEWITHSUNDEEP
pythonmachine-learningclassificationpython-importrandom-forest
user3426614
user3426614

Reputation: 31

VotingClassifier in sklearn.ensemble ImportError

I am trying to implement multiple learning classifiers in python. I have 5 random forest classifiers in the code but now I am not able to import the VotingClassifier function from sklearn.ensemble.

When I write this:

from sklearn.ensemble import VotingClassifier

the error says:

ImportError: cannot import name VotingClassifier

How can I fix this?

Upvotes: 2

Views: 7032

Answers (2)

user2489252
user2489252

Reputation:

I am the person who implemented the VotingClassifier in scikit-learn. Sorry for the confusion, I just stumbled upon the "examples" section in the scikit-learn 16.1 documentation. This is a little bit misleading, the VotingClassifier is already implemented but will be in the next version of scikit-learn 0.17.

If you want to use it "already" you have 2 options:

1) you could install the current scikit-learn dev version available via GitHub: https://github.com/scikit-learn/scikit-learn (instructions are in the Readme)

2) alternatively, you could use it from mlxtend as EnsembleClassifier (http://rasbt.github.io/mlxtend/docs/classifier/scikit-learn_ensemble_classifier/) until the new scikit-version comes out. The mlxtend package is a little "playground" of mine where I upload some examples and functions that I find useful at times.

Hope that helps!

Upvotes: 3

user3666197
user3666197

Reputation: 1

Q: How can I fix this?

Your issue is clear & solve-able. The devil hides in the detail. VotingClassifier was announced in a scikit-learn changelog to be the add-on right for the still-wet-ink release of 0.17.0.

If you are running Anaconda / conda package-manager for python, check:

$ conda search —all scikit-learn and

$ conda depends scikit-learn to verify any newly added dependencies

$ conda create -n (test-0-17-0-sklearn) scikit-learn for creating a new, separate, conda-named / -controlled environment for running python altogether with a sure sklearn ver. 0.17.0 for your further DEV/TEST

A: In any case, wrap the import into a try: { ... } except: { ... }

try:
    from sklearn.ensemble import VotingClassifier
except:
    try:
        import sklearn
        print "WARNING: [VotingClassifier] not available\n",
              "WARNING: [import sklearn] reports version: ",
              sklearn.__version__, "\n"+60*"|"
    except:
        print "WARNING: impossible to [import sklearn] at all\n",
              60*"|"

Upvotes: 0

Related Questions