CODEWITHSUNDEEP
pythonpython-3.xmatrix
Mustafa Aktaş
Mustafa Aktaş

Reputation: 103

How to find row-echelon matrix form (not reduced) in Python?

I am working on a project for my Linear Algebra class. I am stuck with a small problem. I could not find any method for finding the row-echelon matrix form (not reduced) in Python (not MATLAB). Could someone help me out? Thank you.

(I use python3.x)

Upvotes: 3

Views: 4212

Answers (2)

Erik Parkinson
Erik Parkinson

Reputation: 356

Bill M's answer is correct. When you find the LU decomposition the U matrix is a correct way of writing M in REF (note that REF is not unique so there are multiple possible ways to write it). To see why, remember that the LU decomposition finds P,L,U such that PLU = M. When L is full rank we can write this as U = (PL)-1M. So (PL)-1 defines the row operations that you have to preform on M to change it into U.

Upvotes: 3

Bill M.
Bill M.

Reputation: 1548

I think scipy.linalg's "lu" can do that:

>>> from scipy.linalg import lu
>>> import numpy as np
>>> M = np.array([[0,3,-6,6,4,-5], [3,-7,8,-5,8,9], [3,-9,12,-9,6,15]])
>>> p,l,u = lu(M)
>>> u
array([[ 3.        , -7.        ,  8.        , -5.        ,  8.        ,  9.        ],
       [ 0.        ,  3.        , -6.        ,  6.        ,  4.        , -5.        ],
       [ 0.        ,  0.        ,  0.        ,  0.        ,  0.66666667, 2.66666667]])
>>>

Upvotes: 2

Related Questions