How to gracefully create a matrix with the row and column indices as the first row / column?
I am having difficulty creating a matrix for a specific number of rows and columns, a matrix that contains indices as the first row or column respectively.
# At first I create list of lists with 0 at every position
string1 = "dog"
string2 = "hippo"
Dist = [[0 for column in range(len(string1)+1)] for row in range(len(string2)+1)]
I would like to simplify this step if possible:
Dist[0] = [x for x in range(len(string1)+1)]
for x in range(len(string2)+1):
Dist[x][0] = x
This is what the desired and current result looks like (this is a list of lists):
print(Dist)
[[0, 1, 2, 3],
[1, 0, 0, 0],
[2, 0, 0, 0],
[3, 0, 0, 0],
[4, 0, 0, 0],
[5, 0, 0, 0]]
I am planning to use this matrix in a distance calculation problem, but this is a separate part that I don't need to tackle.
My main question is if I am doing it right (I think not) and how to do it better. Anyone, even general advice would be appreciated.
source to share
As you noted numpy
, there is an option here numpy
:
n_cols, n_rows = len(string1)+1, len(string2)+1
Dist = np.zeros((n_rows, n_cols), dtype=np.int32)
Dist[0,:] = np.arange(n_cols)
Dist[:,0] = np.arange(n_rows)
Dist
#array([[0, 1, 2, 3],
# [1, 0, 0, 0],
# [2, 0, 0, 0],
# [3, 0, 0, 0],
# [4, 0, 0, 0],
# [5, 0, 0, 0]], dtype=int32)
source to share
How to simply assign to an np.arange
array np.zero
:
>>> import numpy as np
>>> x = 6
>>> y = 4
>>> arr = np.zeros((x, y), dtype=int)
>>> arr[0, :] = np.arange(y)
>>> arr[:, 0] = np.arange(x)
>>> arr
array([[0, 1, 2, 3],
[1, 0, 0, 0],
[2, 0, 0, 0],
[3, 0, 0, 0],
[4, 0, 0, 0],
[5, 0, 0, 0]])
source to share
Since we're aiming for elegance, here's one compact version of c np.ogrid
that sets the ranges of the arrays for us and then we can assign both of those first rows and columns in one step -
L1,L2 = len(string1)+1, len(string2)+1
Dist1 = np.zeros((L2,L1),dtype=int)
Dist1[:,[0]], Dist1[0] = np.ogrid[:L2,:L1]
Sample output -
In [76]: Dist1
Out[76]:
array([[0, 1, 2, 3],
[1, 0, 0, 0],
[2, 0, 0, 0],
[3, 0, 0, 0],
[4, 0, 0, 0],
[5, 0, 0, 0]])
source to share