sksparse.cholmod-断言错误

当我试图用cholmod分解CSC格式的大型稀疏矩阵时,我得到了一个断言错误。

以下是一个简化的示例:

import numpy as np
from scipy.sparse import coo_array
from sksparse.cholmod import cholesky

row = np.array([0, 1])
col = np.array([0, 1])
val = np.array([1., 1.])


A = coo_array((val, (row, col)), shape=(2, 2)).tocsc()

factor = cholesky(A)

这给了我以下错误:

/tmp/ipykernel_59957/2822312894.py:12: CholmodTypeConversionWarning: converting matrix of class csc_array to CSC format
  factor = cholesky(A)

---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
Cell In[17], line 12
      7 val = np.array([1., 1.])
     10 A = coo_array((val, (row, col)), shape=(2, 2)).tocsc()
---> 12 factor = cholesky(A)

File sksparse/cholmod.pyx:1189, in sksparse.cholmod.cholesky()

File sksparse/cholmod.pyx:1216, in sksparse.cholmod._cholesky()

File sksparse/cholmod.pyx:1126, in sksparse.cholmod._analyze()

File sksparse/cholmod.pyx:411, in sksparse.cholmod._check_for_csc()

AssertionError: 

有人能告诉我我在这里做错了什么吗?

顺便说一句,按以下方式操作还可以:

import scipy.sparse as scsp
factor = cholesky(scsp.csc_matrix(np.array([[ 1., 0.],
       [0.,  1.]]), dtype=np.float64))

但这对于我的非常大的稀疏矩阵来说是低效的。