How to convert a column or row matrix to a diagonal matrix in Python?
80,970
Solution 1
You can use diag method:
import numpy as np
a = np.array([1,2,3,4])
d = np.diag(a)
# or simpler: d = np.diag([1,2,3,4])
print(d)
Results in:
[[1 0 0 0]
[0 2 0 0]
[0 0 3 0]
[0 0 0 4]]
If you have a row vector, you can do this:
a = np.array([[1, 2, 3, 4]])
d = np.diag(a[0])
Results in:
[[1 0 0 0]
[0 2 0 0]
[0 0 3 0]
[0 0 0 4]]
For the given matrix in the question:
import numpy as np
a = np.matrix([1,2,3,4])
d = np.diag(a.A1)
print (d)
Result is again:
[[1 0 0 0]
[0 2 0 0]
[0 0 3 0]
[0 0 0 4]]
Solution 2
I suppose you could also use diagflat:
import numpy
a = np.matrix([1,2,3,4])
d = np.diagflat(a)
print (d)
Which like the diag method results in
[[1 0 0 0]
[0 2 0 0]
[0 0 3 0]
[0 0 0 4]]
but there's no need for flattening with .A1
Solution 3
Another solution could be:
import numpy as np
a = np.array([1,2,3,4])
d = a * np.identity(len(a))
As for performances for the various answers here, I get with timeit
on 100000 repetitions:
-
np.array
andnp.diag
(Marcin's answer): 2.18E-02 s -
np.array
andnp.identity
(this answer): 6.12E-01 s -
np.matrix
andnp.diagflat
(Bokee's answer): 1.00E-00 s
Author by
Tom Kurushingal
Updated on January 27, 2020Comments
-
Tom Kurushingal over 4 years
I have a row vector A, A = [a1 a2 a3 ..... an] and I would like to create a diagonal matrix, B = diag(a1, a2, a3, ....., an) with the elements of this row vector. How can this be done in Python?
UPDATE
This is the code to illustrate the problem:
import numpy as np a = np.matrix([1,2,3,4]) d = np.diag(a) print (d)
the output of this code is [1], but my desired output is:
[[1 0 0 0] [0 2 0 0] [0 0 3 0] [0 0 0 4]]