Strategy 1. sklearn.preprocessing.Imputer
Personally, I do not like this method, because it only offers strategies like mean, median and most frequent. It will add considerable biases to the original dataset and more scientific method should be used!
However, it is the most simple way you can use. Here is the link of this method:
http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.Imputer.html
Strategy 2. fancyimpute in Python
待补充