如果满足值条件，则在numpy数组中将值转换为np.nan

Question

I'm working with a dataset where non existent values show up as a negative number. I want to convert these values to np.nan values but I can't figure out how. The condition for this is (array < 0) .

An example of what would happen to the array would be:

import numpy as np

array = np.array([[-1,  1, -1,  1],
                  [ 1, -1, -1,  1],
                  [ 1, -1, -1, -1]])

To then be converted to:

np.array([[np.nan,      1, np.nan,      1],
          [     1, np.nan, np.nan,      1],
          [     1, np.nan, np.nan, np.nan]])

Cheers

Answer 1

np.nan is a float so you need to convert array to float before doing the boolean masking .

isinstance(np.nan, float)  # True

array = array.astype(float)
array[array < 0] = np.nan
array

outputs

array([[nan,  1., nan,  1.],
       [ 1., nan, nan,  1.],
       [ 1., nan, nan, nan]])