Compare multiple lists and return most frequently occurring number column-wise (most common value by index)

Question

I have three lists:

list1 = [1, 5, 5, 4, 2]
list2 = [1, 4, 5, 3, 4]
list3 = [2, 5, 4, 3, 4]

I want to return the most occurring number in each column so for example:

For these lists, I want to get [1, 5, 5, 3, 4].

The list will always have equal length.

What if there are no 'most occurring' numbers? Are the lists always of equal length? — zwer, Apr 12 '18 at 20:27
Can you show what you've tried? Right now you are basically asking us to do your "assignment" — MooingRawr, Apr 12 '18 at 20:27
@zwer yes the list will always have equal length. If there are no recurring number, I need to compare the accuracy of each list (accuracy is produced based on three types of machine learning algorithm from sci-kit library) and choose the number in the column with the highest accuracy. — Fran, Apr 12 '18 at 20:29

score 2 · Accepted Answer · answered Apr 12 '18 at 20:34

Here's a solution using scipy.stats.mode:

from scipy.stats import mode
m = mode([list1, list2, list3])
print(m)
#ModeResult(mode=array([[1, 5, 5, 3, 4]]), count=array([[2, 2, 2, 2, 2]]))

And if you want just the most frequent values as a list:

print(list(m.mode[0]))
#[1, 5, 5, 3, 4]

1 Answers1