Transpose a matrix in Python [duplicate]

Question

I'm trying to create a matrix transpose function in Python. A matrix is a two dimensional array, represented as a list of lists of integers. For example, the following is a 2X3 matrix (meaning the height of the matrix is 2 and the width is 3):

A=[[1, 2, 3],
   [4, 5, 6]]

To be transposed the jth item in the ith index should become the ith item in the jth index. Here's how the above sample would look transposed:

>>> transpose([[1, 2, 3],
               [4, 5, 6]])
[[1, 4],
[2, 5],
[3, 6]]
>>> transpose([[1, 2],
               [3, 4]])
[[1, 3],
[2, 4]]

How can I do this?

You can find an answer here: stackoverflow.com/questions/6473679/… — 1man
– 1man, Commented Aug 7, 2016 at 14:56

Ashwini Chaudhary · Accepted Answer · 2014-05-10 22:26:57Z

110

You can use zip with * to get transpose of a matrix:

>>> A = [[ 1, 2, 3],[ 4, 5, 6]]
>>> zip(*A)
[(1, 4), (2, 5), (3, 6)]
>>> lis  = [[1,2,3], 
... [4,5,6],
... [7,8,9]]
>>> zip(*lis)
[(1, 4, 7), (2, 5, 8), (3, 6, 9)]

If you want the returned list to be a list of lists:

>>> [list(x) for x in zip(*lis)]
[[1, 4, 7], [2, 5, 8], [3, 6, 9]]
#or
>>> map(list, zip(*lis))
[[1, 4, 7], [2, 5, 8], [3, 6, 9]]

edited May 10, 2014 at 22:26

answered Jun 11, 2013 at 6:13

Ashwini Chaudhary

252k60 gold badges478 silver badges519 bronze badges

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

2 Comments

PM 2Ring Over a year ago

In Python 3, map returns an iterator, so to get a list from the last solution you need list(map(list, zip(*lis))) or [*map(list, zip(*lis))].

kaushalpranav Over a year ago

Can you explain more on how the * works in zip(*lis)

Ashwini Chaudhary · Accepted Answer · 2014-04-24 21:14:40Z

28

Is there a prize for being lazy and using the transpose function of NumPy arrays? ;)

import numpy as np

a = np.array([(1,2,3), (4,5,6)])

b = a.transpose()

edited Apr 24, 2014 at 21:14

Ashwini Chaudhary

252k60 gold badges478 silver badges519 bronze badges

answered Jun 11, 2013 at 10:54

Klaus-Dieter Warzecha

2,3652 gold badges28 silver badges34 bronze badges

2 Comments

Asher G. Over a year ago

Your answer is not correct because of the given skeleton function. This answer must be done using list comprehension. The answer below is correct. Good information though.

Ehsan Kia Over a year ago

Not only you're importing a huge library to do something that can be achieved with one line of pure python, your answer is also technically wrong because the output is not a list of list as shown in the question, but a numpy array. So the answer is longer and slower, because you have to convert to a numpy array, transpose, and convert back.

Asher G. · Accepted Answer · 2013-06-11 06:11:13Z

13

If we wanted to return the same matrix we would write:

return [[ m[row][col] for col in range(0,width) ] for row in range(0,height) ]

What this does is it iterates over a matrix m by going through each row and returning each element in each column. So the order would be like:

[[1,2,3],
[4,5,6],
[7,8,9]]

Now for question 3, we instead want to go column by column, returning each element in each row. So the order would be like:

[[1,4,7],
[2,5,8],
[3,6,9]]

Therefore just switch the order in which we iterate:

return [[ m[row][col] for row in range(0,height) ] for col in range(0,width) ]

answered Jun 11, 2013 at 6:11

Asher G.

5,0115 gold badges31 silver badges30 bronze badges

Collectives™ on Stack Overflow

Transpose a matrix in Python [duplicate]

3 Answers 3

2 Comments

2 Comments

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

2 Comments

Comments

Linked

Related