与scikit-learn的兼容性问题

Question

我正在尝试在Python2.7上使用scikit-learn，但导入会产生错误。

from sklearn.preprocessing import normalize
import numpy as np

这会产生以下错误

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
<ipython-input-2-1608d2dbd4a8> in <module>()
      2 from scipy.sparse.linalg import svds
      3 from scipy.sparse import hstack, diags, csr_matrix
----> 4 from sklearn.preprocessing import normalize
      5 import numpy as np
      6 import csv

C:\Python27\lib\site-packages\sklearn\preprocessing\__init__.pyc in <module>()
      4 """
      5 
----> 6 from .data import Binarizer
      7 from .data import KernelCenterer
      8 from .data import MinMaxScaler

C:\Python27\lib\site-packages\sklearn\preprocessing\data.pyc in <module>()
     13 from ..base import BaseEstimator, TransformerMixin
     14 from ..externals import six
---> 15 from ..utils import check_arrays
     16 from ..utils import atleast2d_or_csc
     17 from ..utils import array2d

C:\Python27\lib\site-packages\sklearn\utils\__init__.pyc in <module>()
      8 import warnings
      9 
---> 10 from .murmurhash import murmurhash3_32
     11 from .validation import (as_float_array, check_arrays, safe_asarray,
     12                          assert_all_finite, array2d, atleast2d_or_csc,

ImportError: cannot import name murmurhash3_32

我使用pip install -U scikit-learn安装了scikit-learn 有些人可以帮我解决这个问题。