将Matlab代码转换为python(scipy) [英] Translate matlab code to python (scipy)

问题描述

我正在尝试将此matlab代码转换为python:

I' trying to translate this matlab code to python:

T = length(z);
lambda = 10;
I = speye(T)
D2 = spdiags(ones(T-2,1)*[1 -2 1],[0:2],T-2,T);
z_stat = (I-inv(I + lambda^2*D2'*D2))*z;

我现在得到的是:

T = len(signal)
lam = 10;
I  = np.identity(T)
D2 = scipy.sparse.spdiags(np.ones((T-2,1),dtype=np.int)*[1,-2,1],(range(0,3)),T-2,T);

此刻我收到此错误

"scipy.sparse.sp ... ge(0,3))，T-2，T)" ValueError:对角线数 (298)与偏移数不匹配(3)args元组:('number 对角线(298)与偏移数(3)'，)不符

"scipy.sparse.sp...ge(0,3)),T-2,T)" ValueError: number of diagonals (298) does not match the number of offsets (3) args tuple: ('number of diagonals (298) does not match the number of offsets (3)',)

查看文档时，matlab函数和python函数非常相似.虽然可能有一个我缺少的差异.现在的问题是:我在做什么错了?

When looking at the documentation, the matlab function and the python function are very similar. Though there is probably a difference which I am missing. My question is now: What am I doing wrong ?

编辑:z是一个长度为300的数组

edit: z is an array with length 300

推荐答案

如果将过滤器数据转置为spdiags，则在两个程序包中都得到相同尺寸的答案:

If you transpose the filter data to spdiags then you get an answer of the same dimensions in both packages:

# numpy/scipy
filt = [1,-2,1]* np.ones((1,T-2),dtype=np.int).T
D2 = scipy.sparse.spdiags(data.T, (range(0,3)),T-2,T)
np.shape(D2)
>>> (298, 300)

% matlab check
D2 = spdiags(ones(T-2,1)*[1 -2 1],[0:2],T-2,T)
size(D2)
ans =
  298   300

这篇关于将Matlab代码转换为python(scipy)的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！