解决：Input contains NaN, infinity or a value too large for dtype('float64')

　源代码

from sklearn.preprocessing import Imputer
data = Imputer(missing_values = 'NaN',strategy='most_frequent')

　出现错误：

cannot import name 'Imputer' from 'sklearn.preprocessing' (G:\Anaconda\lib\site-packages\sklearn\preprocessing\__init__.py)

　原因分析：是因为Imputer已经不在preprocessing模块下，改变为impute下了

　修改代码如下：

from sklearn.impute import SimpleImputer
data = SimpleImputer(missing_values = 'NaN',strategy='most_frequent')

　　第二个问题：

Input contains NaN, infinity or a value too large for dtype('float64').

　　包含太大的值

修改为：

data = SimpleImputer(missing_values = np.NaN,strategy='most_frequent')

　　问题解决！