My Pandas dataframe shows the number 0 as not null

Question

I would like to delete every row that contains 0 in order to have cleaner data. However as you can see on the image it shows that 0 is not a null value. Can you guys help me on this? Thank you.

Answer 1

As you can read in the pandas doc the method DataFrame.notnull() check if the value are NA, and not for the values equal to zero or in your case a string because you replaced it with 'NA'.

Now to do what you want you can do:

df[(df != 0).all(1)]

As describe in this post

Answer 2

Well, when you are trying df.replace(0,'NA') you are subbing in the string value of NA which is not NULL.

You can just use np.nan there like : df.replace(0,np.nan) [with import numpy as np]

Answer 3

By using a replace, you might want to use the following:

df.replace(0, np.nan).dropna()

It basically replaces zeroes by np.nan wich can be eliminated by pandas' .dropna()