Introduction

This is a Python learning memo.

problem

If there are 0s in the array created by numpy, dividing by 0 will give a warning. (Of course)

`problem.py`


import numpy as np

A = np.array([0, 1, 2])
B = np.array([0, 1, 1])

print(A/B)

output> RuntimeWarning: invalid value encountered in true_divide
  print(A/B)
[nan  1.  2.]

A warning is issued, and the part divided by zero is nan.

Solution

`solved.py`


import numpy as np

A = np.array([0, 1, 2], dtype=float)
B = np.array([0, 1, 1], dtype=float)

C = np.divide(A, B, out=np.zeros_like(A), where=B!=0)
print(C)

output> [0.  1.  2.]

np.zeros_like (A) returns a zero-filled array of the same shape as A. You can also use np.divide (A, B) to divide A by B. According to the documentation, out is the place to save the results and where is the option to specify conditions for all inputs. 0 is assigned to the calculation result of the place where where is false (B [0] in this case).

in conclusion

Sure, the warning is gone, but it can help, so I'm wondering if it's really a good solution.

Thank you for visiting. If you have any suggestions, please leave them in the comments section.

Reference URL

https://stackoverflow.com/questions/26248654/how-to-return-0-with-divide-by-zero (viewed February 24, 2020) https://docs.scipy.org/doc/numpy/reference/generated/numpy.divide.html(2020年2月24日閲覧) https://docs.scipy.org/doc/numpy/reference/generated/numpy.zeros_like.html(2020年2月24日閲覧)

[PYTHON] Avoid the warning [invalid value encountered in true_divide] when dividing by zero