How to we replace log(0) with 0?

Question

RuntimeWarning: invalid value encountered in multiply

I have a code:

a = Y_list * np.log(Y_list/E_Y) print(a)

My Y_list contains 0 values, I'm wondering how to do when Y_list = 0 , np.log(0) = 0?

hi thanks for the reply! im actually doing a school assignment and they require such a condition..'If Y = 0, the expression log[Y/(E(Y))] will be taken as zero' — Rachel L
– Rachel L, Commented Aug 30, 2021 at 12:58
are Y_list and E_Y numpy arrays? you can use np.where or an inline if depending on the answer. — alparslan mimaroğlu
– alparslan mimaroğlu, Commented Aug 30, 2021 at 13:02

alparslan mimaroğlu · Accepted Answer · 2021-08-30 13:54:03Z

3

You can use np.where It lets you define a condition for true and false and assign different values.

np.where((Y_list/E_Y)!= 0, np.log(Y_list/E_Y),0)

answered Aug 30, 2021 at 13:19

1,4901 gold badge14 silver badges22 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

user1717828 · Accepted Answer · 2022-11-04 17:29:46Z

Alternatively, we can run np.log with a where parameter:

import numpy as np a = np.arange(0, 5000, 1000) np.log(a, where=a != 0) # array([0. , 6.90775528, 7.60090246, 8.00636757, 8.29404964])