python3.4 sklearn安装陷阱

1、由于早期用python3.4研究pandas时，顺便安装了numpy，当时安装的是numpy 1.12.0rc2+mkl，而这次我打算自己研究sklearn程序，发现安装了sklearn '0.19.2'后，import sklearn 异常

D:\Python34\Scripts>python
Python 3.4.3 (v3.4.3:9b73f1c3e601, Feb 24 2015, 22:43:06) [MSC v.1600 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import sklearn
Traceback (most recent call last):
  File "__init__.pxd", line 998, in numpy.import_array
RuntimeError: module compiled against API version 0xc but this version of numpy is 0xa

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "", line 1, in
  File "D:\Python34\lib\site-packages\sklearn\__init__.py", line 134, in
    from .base import clone
  File "D:\Python34\lib\site-packages\sklearn\base.py", line 13, in
    from .utils.fixes import signature
  File "D:\Python34\lib\site-packages\sklearn\utils\__init__.py", line 9, in
    from .murmurhash import murmurhash3_32
  File "sklearn\utils\murmurhash.pyx", line 26, in init sklearn.utils.murmurhash
  File "__init__.pxd", line 1000, in numpy.import_array
ImportError: numpy.core.multiarray failed to import
>>> import scipy
>>> import numpy
>>> import sklearn
Traceback (most recent call last):
  File "", line 1, in
  File "D:\Python34\lib\site-packages\sklearn\__init__.py", line 133, in
    from . import __check_build
ImportError: cannot import name '__check_build'

经定位是numpy+mkl版本太低了需要安装numpy-1.14.5+mkl-cp34-cp34m-win32