/usr/local/bin/EPDpython/lib/python2.7/site-packages/scikit_learn-0.12_git-py2.7-linux-i686.egg/sklearn/ensemble/forest.pyc in fit(self, X, y)
    252                 self.random_state.randint(MAX_INT),
    253                 verbose=self.verbose)
--> 254             for i in xrange(n_jobs))
    255 
    256         # Reduce


/usr/local/bin/EPDpython/lib/python2.7/site-packages/scikit_learn-0.12_git-py2.7-linux-i686.egg/sklearn/externals/joblib/parallel.pyc in __call__(self, iterable)
    471         try:
    472             for function, args, kwargs in iterable:
--> 473                 self.dispatch(function, args, kwargs)
    474 
    475             self.retrieve()

/usr/local/bin/EPDpython/lib/python2.7/site-packages/scikit_learn-0.12_git-py2.7-linux-i686.egg/sklearn/externals/joblib/parallel.pyc in dispatch(self, func, args, kwargs)
    294         """
    295         if self._pool is None:
--> 296             job = ImmediateApply(func, args, kwargs)
    297             index = len(self._jobs)
    298             if not _verbosity_filter(index, self.verbose):

/usr/local/bin/EPDpython/lib/python2.7/site-packages/scikit_learn-0.12_git-py2.7-linux-i686.egg/sklearn/externals/joblib/parallel.pyc in __init__(self, func, args, kwargs)
    122         # Don't delay the application, to avoid keeping the input

    123         # arguments in memory

--> 124         self.results = func(*args, **kwargs)
    125 
    126     def get(self):

/usr/local/bin/EPDpython/lib/python2.7/site-packages/scikit_learn-0.12_git-py2.7-linux-i686.egg/sklearn/ensemble/forest.pyc in _parallel_build_trees(n_trees, forest, X, y, sample_mask, X_argsorted, seed, verbose)
     75             indices = random_state.randint(0, n_samples, n_samples)
     76             tree.fit(X[indices], y[indices],
---> 77                      sample_mask=sample_mask, X_argsorted=X_argsorted)
     78             tree.indices_ = indices
     79 

/usr/local/bin/EPDpython/lib/python2.7/site-packages/scikit_learn-0.12_git-py2.7-linux-i686.egg/sklearn/tree/tree.pyc in fit(self, X, y, sample_mask, X_argsorted)
    468             self.classes_ = np.unique(y)
    469             self.n_classes_ = self.classes_.shape[0]
--> 470             criterion = CLASSIFICATION[self.criterion](self.n_classes_)
    471             y = np.searchsorted(self.classes_, y)
    472 

KeyError: None