ImportError: cannot import name VarianceThreshold

Question

scikit-learn seems to work, but when I did:

from sklearn.feature_selection import VarianceThreshold

I got the following error:

ImportError: cannot import name VarianceThreshold

How to bypass this? I am a newbie in Python, so I have no idea what to do.

I played with the order of my imports, as suggested here: ImportError: Cannot import name X, but no luck.

import sys
import pandas as pd
import numpy as np
import operator
from sklearn.feature_selection import VarianceThreshold
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.feature_extraction.text import HashingVectorizer
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.preprocessing import normalize
from sklearn import decomposition

---

I am also getting this:

code/python/k_means/serial_version$ python -c 'import sklearn; print(sklearn.VarianceThreshold)'
Traceback (most recent call last):
  File "<string>", line 1, in <module>
AttributeError: 'module' object has no attribute 'VarianceThreshold'

---

Version:

>>> import sklearn
>>> sklearn.__version__
'0.14.1'

Answer 1

You can bypass by catching the exception

try:
    from sklearn.feature_selection import VarianceThreshold
except:
    pass  # it will catch any exception here

If you want to catch only Attribue Error Exception then use below

try:
    from sklearn.feature_selection import VarianceThreshold
except AttributeError:
    pass # catches only Attribute Exception