CODEWITHSUNDEEP
pythonlist-comprehension
Jeremy
Jeremy

Reputation: 2000

List comprehension to normalize values in a matrix

I have a matrix of values, like:

matrix = [
[1,2,3],
[5,6,7]
]

I want to normalize them, so that each row sums to one. This is pretty simple with an approach like:

result = []
for x in matrix:
    curr_row = [z/sum(x) for z in x]
    result.append(curr_row)

I'm wondering if there's a way to do this with a list comprehension.

The closest I've gotten is

result = [x/sum(y) for y in matrix for x in y]

but this collapses it all into one list.

Upvotes: 1

Views: 113

Answers (1)

SparkAndShine
SparkAndShine

Reputation: 18027

You are almost there.

from __future__ import division        # for Python2

results = [[item/sum(row) for item in row] for row in matrix] # a list of lists

print(results)
# Output
[[0.16666666666666666, 0.3333333333333333, 0.5], [0.2777777777777778, 0.3333333333333333, 0.3888888888888889]]

Upvotes: 2

Related Questions