如何在SciPy中创建对角稀疏矩阵
问题描述：
我正在尝试创建一个稀疏矩阵，该矩阵具有沿对角线排列的2D模式.举个简单的例子，这可能最容易解释.
I am trying to create a sparse matrix which has a 2D pattern run down the diagonal. This is probably easiest to explain with a quick example.
说我的模式是:[1,0,2,0,1] ...
Say my pattern is: [1,0,2,0,1]...
我想创建一个稀疏矩阵:
I want to create a sparse matrix:
[[2,0,1,0,0,0,0...0],
[0,2,0,1,0,0,0...0],
[1,0,2,0,1,0,0...0],
[0,1,0,2,0,1,0...0],
[0,0,1,0,2,0,1...0],
[...]]
scipy.sparse.dia_matrix似乎是一个不错的选择，但是，我根本无法从现有文档中弄清楚如何完成我想要的工作.预先谢谢你
The scipy.sparse.dia_matrix seems like a good candidate, however, I simply cannot figure out how to accomplish what I want from the documentation available. Thank you in advance
答
N = 10
diag = np.zeros(N) + 2
udiag = np.zeros(N) + 1
ldiag = np.zeros(N) + 1
mat = scipy.sparse.dia_matrix(([diag, udiag, ldiag], [0, 2, -2]), shape=(N, N))
print mat.todense()
[[ 2. 0. 1. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 2. 0. 1. 0. 0. 0. 0. 0. 0.]
[ 1. 0. 2. 0. 1. 0. 0. 0. 0. 0.]
[ 0. 1. 0. 2. 0. 1. 0. 0. 0. 0.]
[ 0. 0. 1. 0. 2. 0. 1. 0. 0. 0.]
[ 0. 0. 0. 1. 0. 2. 0. 1. 0. 0.]
[ 0. 0. 0. 0. 1. 0. 2. 0. 1. 0.]
[ 0. 0. 0. 0. 0. 1. 0. 2. 0. 1.]
[ 0. 0. 0. 0. 0. 0. 1. 0. 2. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 1. 0. 2.]]