使用Rodeo IDE时无法导入Python“ statsmodels”

Question

我先安装了Anaconda，然后安装了Rodeo（使用其网站上的安装文件）。 但是，当我运行脚本时，出现以下错误。 问题所在是：

import statsmodels.api as sm

奇怪的是，此代码使用Spyder2 IDE可以完美运行。

有什么建议么？

史蒂夫

>>> import statsmodels.api as sm
---------------------------------------------------------------------------
NotImplementedError                       Traceback (most recent call last)
<ipython-input-10-6030a6549dc0> in <module>()
----> 1 import statsmodels.api as sm

C:\Users\steve\Anaconda2\lib\site-packages\statsmodels\__init__.py in <module>()
      6 
      7 from warnings import simplefilter
----> 8 from .tools.sm_exceptions import (ConvergenceWarning, CacheWriteWarning,
      9                                   IterationLimitWarning, InvalidTestWarning)
     10 

C:\Users\steve\Anaconda2\lib\site-packages\statsmodels\tools\__init__.py in <module>()
----> 1 from .tools import add_constant, categorical

C:\Users\steve\Anaconda2\lib\site-packages\statsmodels\tools\tools.py in <module>()
      7 import numpy.linalg as L
      8 from scipy.linalg import svdvals
----> 9 from statsmodels.distributions import (ECDF, monotone_fn_inverter,
     10                                        StepFunction)
     11 from statsmodels.datasets import webuse

C:\Users\steve\Anaconda2\lib\site-packages\statsmodels\distributions\__init__.py in <module>()
----> 1 from .empirical_distribution import ECDF, monotone_fn_inverter, StepFunction
      2 from .edgeworth import ExpandedNormal
      3 

C:\Users\steve\Anaconda2\lib\site-packages\statsmodels\distributions\empirical_distribution.py in <module>()
      3 """
      4 import numpy as np
----> 5 from scipy.interpolate import interp1d
      6 
      7 def _conf_set(F, alpha=.05):

C:\Users\steve\Anaconda2\lib\site-packages\scipy\interpolate\__init__.py in <module>()
    143 from __future__ import division, print_function, absolute_import
    144 
--> 145 from .interpolate import *
    146 from .fitpack import *
    147 

C:\Users\steve\Anaconda2\lib\site-packages\scipy\interpolate\interpolate.py in <module>()
     29 from . import _ppoly
     30 from .fitpack2 import RectBivariateSpline
---> 31 from .interpnd import _ndim_coords_from_arrays
     32 
     33 

scipy\interpolate\interpnd.pyx in init scipy.interpolate.interpnd (scipy\interpolate\interpnd.c:24330)()

C:\Users\steve\Anaconda2\lib\site-packages\scipy\spatial\__init__.py in <module>()
     89 
     90 from .kdtree import *
---> 91 from .ckdtree import *
     92 from .qhull import *
     93 from ._plotutils import *

ckdtree.pyx in init ckdtree (scipy\spatial\ckdtree\ckdtree.cxx:22661)()

C:\Users\steve\Anaconda2\lib\multiprocessing\__init__.pyc in cpu_count()
    134         return num
    135     else:
--> 136         raise NotImplementedError('cannot determine number of cpus')
    137 
    138 def freeze_support():

NotImplementedError: cannot determine number of cpus

Answer 1

scipy中的多处理库有一个依存关系，用于寻找您可用的内核数。 我们已在最新版本中添加了对此功能的检测，但是与此同时，您可以手动设置环境变量：

import os
os.environ["NUMBER_OF_PROCESSORS"] = "2"